FYI Technical Site News (for those interested)

[gerryvz]

Moderator's Note:

I understand that many if not most members don't give a rip as to what underpins the site and how things work.

For those few folks who ARE interested in how the site works, I created a new thread that will be dedicated to the more "technical" aspects of the site, and its ongoing evolution/updates.

I will continue to update this thread on an ongoing basis with a more technical description of any changes or updates to things. Folks are welcome to chime in with questions, suggestions, corrections, advice, or comments as they wish.

===============================


I'm likely going to do a Linux system update this coming weekend, probably early on Saturday morning. This is to update a number of the software packages that underlie the forum. These are mainly minor security updates to about a dozen Linux programs.

There shouldn't be any down-time associated with the update, but I will likely take the forum off-line for approximately 5-10 minutes while the server software is being downloaded and updated, for safety purposes. It will also probably entail a system reboot, which takes about 90 seconds as part of that 5-10 minutes.

Going forward, this will probably be a monthly to quarterly thing I'll be doing, as I have to maintain and administer the Linux system software myself .... a step I didn't have to do in the past with the old forum.

Cheers,
Gerry

 

[gerryvz]

I wanted to let folks know that I will be taking the redirect system down for a few hours, likely tomorrow (Thursday) morning from about 5-8 AM Eastern Time. This is to do some further migration of remaining photo albums from the old site over to this site.

During this period, links to posts on the old forum will NOT redirect to the appropriate post on this site. I am trying to do this at a time when few people will be using this site, so the impact should be minimal. ALL content existing on this site will continue to be available, you can post/reply, and so forth, during this time.

Thanks for your patience while I make this final push to get the Photo Albums migrated.

Cheers,
Gerry




I have 53 photo albums yet to migrate or delete. I believe up to two-thirds of those albums were deleted by their owners, but I need to double-check them (one at a time) against the records on the other site.

 

[gerryvz]

I just rebooted the system to apply around 27 security patches and updates to the Linux kernel. The forum was down for about 1 minute in total. The up-time was 18 days and 23 hours since the last system reboot, FYI. Just normal system maintenance today.

Cheers,
Gerry

 

[gerryvz]

Over the next couple of weeks, I will be updating the version of Java (including the Java Development Kit, Java runtime environment and Java compiler) on the server that runs the Elasticsearch search engine. This is because the installed version of Java, which is called Java 8 Standard Edition (SE), is approaching "end of life" status. My employer, Oracle Corp., late last year released the next "long term support" version of Java, which is called Java 11.

Java 11 will be actively supported for at least the next five years, and the version of Elasticsearch that I have installed and running already supports Java 11. Java 11 offers some significant performance and functionality improvements over Java 8.

Here are Oracle's official release notes pertaining to Java 11: https://www.oracle.com/technetwork/java/javase/11-relnote-issues-5012449.html

Cheers,
Gerry

 

[gerryvz]

Today, XenForo released version 2.1.2 of the forum software, along with an updated version of the Media Gallery software.

I will be updating these in the near future. This is mainly a bug-fix version of the software, so there should be few new features and enhancements.



The new versions of XenForo and the Media have been installed and the upgrade has been completed.

Also updated 26 Ubuntu packages, 16 of which were security updates.
This also incrementally udpated the version of MySQL from 5.7.25 to 5.7.26

Next system updates will be to (as mentioned above) update Java from version 8 to version 11.
I will also update php from 7.2.17 to 7.3.

 

[JoeB]

Gerry, are you running this on a VPS or cloud/shared hosting?
I am guessing the former...
I noticed your av in the XF forum. I would be keen to know how much of a PITA it was to convert vb4 over.
we're still running vb on 190rev and phpbb on ozbenz (which just had a huge provider migration outage) so looking for alternatives and "less" hassle - if there is such a thing.

 

[gerryvz]

Gerry, are you running this on a VPS or cloud/shared hosting?
I am guessing the former...
I noticed your av in the XF forum. I would be keen to know how much of a PITA it was to convert vb4 over.
we're still running vb on 190rev and phpbb on ozbenz (which just had a huge provider migration outage) so looking for alternatives and "less" hassle - if there is such a thing.

I went from a managed/shared hosting account with 1and1.com (now IONOS), which was pretty reliable overall (I used them since 2012, and only had a couple of day-or-two-long outages over that time), to a Virtual Private Server with Linode, which has been excellent.

I had done much research into VPS providers, of which there are many. I had come across Linode many times as a part of a well-respected group of them. However, I was at a work-related conference last December in Seattle, and Linode had a small booth there. I talked to them at the booth and told them they were on my short-list and what I wanted to do in terms of checking out their service as a migration option. They immediately (as part of their show promotion) gave me a free $50 credit, and encouraged me to sign up and create my account. I didn't do this until February 2019, and started delving into it. I was impressed at the depth and extent of their documentation and assistance with setup and configuration. I have used their tech support a couple of times and they have been SUPER responsive, and in one case where I created a new storage volume (and hadn't meant to), they immediately gave me, unasked, a $10 credit for something that I did, and they easily helped fix, because they felt that their documentation may have been slightly confusing.

Linode has data centers all over the world. I am hosting this site out of their Newark, NJ data center. For you in APAC, they have data centers in both Tokyo and Singapore, but nothing in OZ (yet). I am paying by the month, and currently, hosting costs are running me around $45-50 a month. I will be reducing this to around $25 a month very soon.

There is plenty of documentation about converting vBulletin forums over to XenForo. XenForo's vBullet in importers are quite good and complete, and they do revise them from time to time. Their importers work for vB versions 3.X, 4.X and 5.X. I have contributed to their knowledge base a bit, based on my experiences here. There are also quite a number of good threads from people who have converted over their forums, and the guide in the XenForo documentation is also quite good, at least as a general guide and starting point.

You can certainly purchase a XenForo license and download and install it, and you can unlimited "test" forums (but only one "real" forum per license). If you don't like their software, you can get a full refund from them. Again, XenForo tech support has been quite responsive and helpful to me.

I first started looking at XenForo back in 2014, as you can see from the "conversion to XenForo" thread here. At that time they were on version 1.3 or 1.4 of their forum software. It was quite immature at that time. Version 1.5 was a major inflection point in the maturity of the platform, as was the 2.0 betas that started in 2016, with the final 2.0 code being finalized in 2017, I believe. Version 2.1, believe it or not, was also a major step forward in the maturity of the platform, and it was officially released around the end of 2018 (after a beta period during 2018). We are now on the third "point" release of 2.1, which was just released this week, version 2.1.2, a bug fix version. 2.1 added a lot of key capabilities, including their emoji handling and "likes" system which was a big deal for a lot of people.

I'd say that a decent sized forum could be imported and moved over probably in 24-36 hours with little difficulty. I recommend doing AT LEAST three "test" imports with "virgin" XenForo forum software installs before trying a "real" migration, just to get the process down. I did this and even then had a few niggling issues (which I documented a bit here during the weekend I migrated).

I also HIGHLY recommend going to a VPS vs. a hosted provider. The main reason for this is that you have MUCH MUCH MUCH more flexibility as to forum settings, for example database sizes, inode (file) count limitations (1&1 had a 250K inode count limit, no exceptions, vs. 10+ million inodes available with a VPS), and ability to install non-forum software such as ImageMagick, ElasticSearch, and other packages that really help enhance the forum's back-end.

The down-side to a VPS is that you are TOTALLY responsible for system updates, security patches to the operating system, and so forth. It is exceedingly easy to mess things up if you don't know what you are doing or are inexperienced at using Linux. Security is REALLY IMPORTANT at multiple levels with a VPS, whereas with a forum all you basically need to worry about is the forum software itself, as well as your directories on the hosted server.

phpBB is crap. It's free, and it's basic and functional, and there are SOME mods available for it, but that's about the best thing I can say about it. You get what you pay for. vBulletin (even the current version 5) IMHO is also crap. They really dropped the ball with version 4, which became a huge hairball/kludge of code, cobbled together that worked ... but was creaky. 4.X was in a lot of ways a down-step from 3.X, simply because 4.X was not created and maintained by the folks who did the original 3.X code (logically! The original vBulletin guys left and went on to found XenForo). 5.X is their attempt to get things back on track with a more modern forum software architecture, but judging by the numbers of people who are migrating off of that platform to others such as XenForo, it does not seem to be igniting the world on fire with love for it. I believe XenForo also has importers for phpBB as well.

Since migrating over in early March (I think around March 13 to be exact), I have been spending 3-5 hours per day tweaking this site, making modifications and additions, and generally bringing it up to a level of maturity and stability that members here deserve and expect. There is still a lot of work to do, but it is ongoing. I am hoping to be able to take the "old" forum offline (it's in accessible to all now) within the next 30-60 days, as soon as the Wiki software is stable and installable here, and when the Wiki content can be migrated off of the old site to this one.

Bottom line, is that yes I would recommend XenForo over vBulletin. It's robustly supported, actively developed, and has a great community of developers who create cool stuff. If you need help with a migration, let me know. I'd be happy to give you the benefit of my own experience, pointers and some ideas to make the process easier.

You can install XenForo on a hosted server no problem, and it works great. Some of the decision depends on how large and active your forum is, and what you want to do with it. XenForo's admin tools are easlier and far better than vBulletin's, and overall I don't think a XenForo forum on a managed/hosted server would be MORE time/effort than vBulletin, and very likely somewhat less after iniitial migration and tweaking.

Cheers,
Gerry

 

[gerryvz]

A small dot-revision in the elasticsearch software (from 6.7.1 to 6.7.2) was released today, so I did a quick and seamless upgrade.

I am not quite ready to upgrade to the very recently released 7.0.0 version.

 

[gerryvz]

Sorry for the four-hour site outage today. That was my fault. I upgraded the php language from 7.2 to 7.3, and it didn't go well due to configuration and a couple of php extension files that didn't get installed in the upgrade. I decided to revert back to 7.2 (which works perfectly fine), but in the process I had to re-add the same php extension files and then reconfigure everything from scratch.

In the middle of all that, I had to leave for a couple of hours.

I also installed today a new version of phpMyAdmin, which is the software program that helps administer (back up, mostly) the forum's database. This is a newer version, version 4.8.5 (dating from January, 2019), upgrading from the previous version 4.6.5 (which dates from December, 2016).

Again, apologies to all for the four-hour outage. I'll try to do better in the future and keep the outages to less time.

Cheers,
Gerry

 

[gerryvz]

The VaultWiki software has now moved into "Beta" stage, from numerous "Alpha" versions, so I am feeling solid enough about it that I will install it on a test basis, and work with the developer to do an import of the info from the old vBulletin forum.

I expect that this install will happen within the next five days. There won't be any disruption to the forum while I'm doing this. When things are done and installed, I'll post an "official" update for everyone.

Cheers,
Gerry

 

[gerryvz]

I rebooted the server just now, after 11 days' uptime, because I installed a number of security patches and updates to the Linux core system files. Pretty routine, but important maintenance to keep up with.

Total site down-time was around 1.5 minutes.

The following NEW packages will be installed:
linux-headers-4.15.0-50 linux-headers-4.15.0-50-generic
linux-image-4.15.0-50-generic linux-modules-4.15.0-50-generic
linux-modules-extra-4.15.0-50-generic

The following packages will be upgraded:
linux-generic linux-headers-generic linux-image-generic linux-libc-dev

 

[gerryvz]

Did a weekly system update with new files. I am probably going to take the forum down for a few minutes today, to migrate from the current php 7.2 version, to php 7.3. This should be a fairly minor switchover, as I believe I have all of the required additional modules for php 7.3 already uploaded to the server and activated. If there is a snag, I don't think there would be more than about 15 minutes of downtime, while I download and install anything that I missed for the upgrade.

Cheers,
Gerry

The following items were changed:

libapache2-mod-php7.2
libapache2-mod-php7.3
libseccomp2 libssl1.1
openssl
php7.0-common
php7.0-gd
php7.2
php7.2-bcmath
php7.2-bz2
php7.2-cli
php7.2-common
php7.2-curl
php7.2-gd
php7.2-gmp
php7.2-intl
php7.2-json
php7.2-mbstring
php7.2-mysql
php7.2-opcache
php7.2-readline
php7.2-xml
php7.3
php7.3-bcmath
php7.3-bz2
php7.3-cli
php7.3-common
php7.3-curl
php7.3-fpm
php7.3-gd
php7.3-gmp
php7.3-intl
php7.3-json
php7.3-mbstring
php7.3-mysql
php7.3-opcache
php7.3-readline
php7.3-recode
php7.3-tidy
php7.3-xml
php7.3-xmlrpc
php7.3-zip
python3-software-properties
software-properties-common

 

[gerryvz]

I just took the board off-line for a total of two minutes. This was to switch the version of php (the programming language that the forum software uses) from the previous version 7.2.19, to the current version 7.3.6.

Purpose is to keep things current. I will be updating a few more items, including the ElasticSearch enging, the Apache web server software, and the MySQL database over the next couple of weeks, but nothing immediate as I am traveling solidly for the next 8 days.

 

[gerryvz]

I just issued a full system reboot, which resulted in the forum being down & unavailable for approximately 1 minute.

This was due to installing 19 updates to the Ubuntu Linux software. 10 of these updates were security patches, and 9 of the files were updates to the core Linux files.

I also took the opportunity to update the Ubuntu Linux kernel from version 4.15 to the current 4.18, and its associated new hardware enablement stack (HWE), which helps make future updates easier.

The last full system reboot was 24 days and 20 hours (almost 25 days) ago. Full reboots are not often required, unless the core system files are updated. Normally I can leave the system up and running when doing system maintenance, and I can stop and re-start individual services such as the Apache web server, MySQL database or php language with very limited (<5 seconds) disruption to the system/forum operation.

Generally, Ubuntu is updated on a near-daily basis, so I try to group the major system updates to about a once-per-month cadence, so that system downtime (even ~1 minute per month) is reduced to a minimum.

Occasionally when larger system updates are required (such as when I recently moved from php 7.2 to 7.3, or a major Linux version change (next April we will move from Ubuntu 18.04.x to 20.04.x) the system can be down for up to 10-15 minutes.

Cheers from Orlando, FL,
Gerry

 

[gerryvz]

Updated the version of the Apache2 web server software from version 2.4.29 to the current version 2.4.39 while waiting at the gate to depart Orlando back to Baltimore.

 

[LWB250]

You better get out in a hurry, there's a lot of nasty weather coming your way....

Dan

 

[gerryvz]

Sitting on the runway for the past two hours & counting ....

 

[gerryvz]

I just received the following note from the folks at Linode, our hosting provider:

Hello,

Maintenance is required for one or more of your Linodes' physical hosts. This maintenance is necessary to patch the MDS (ZombieLoad) CPU vulnerability[1] and will incur downtime for your Linode(s).

Some FAQ:

- Your maintenance schedule can be found at the bottom of this ticket. The scheduled start times are also listed within the Linode Manager[2/3].
- These maintenance windows cannot be altered, postponed, or rescheduled, and rebooting your Linode ahead of time will not remove the need for this maintenance.
- A two-hour window is allocated, however, the actual downtime should be much less.
- Your Linode(s) will be cleanly shutdown and will remain inaccessible during the maintenance window. After the maintenance has concluded, each Linode will be returned to its last state (running or powered off).
- These updates affect the underlying infrastructure that your Linode(s) reside on and will not affect the data stored within your Linode.

To fully mitigate the MDS vulnerability, set your Configuration Profile to utilize our latest kernel. If you use a distribution-supplied kernel, you can update your kernel[4] via your package manager. Linodes that use our latest kernel will automatically update upon reboot.

To ensure your Linode's essential services (MySQL, Apache, NGINX, etc.) return successfully upon reboot, we strongly recommend that you review our Reboot Survival Guide[5] prior to the start of the maintenance period. If you encounter any connectivity issues after the maintenance is complete, please refer to this Community Site post[6].

Please let us know if you have any questions or concerns. Thanks!


This means that the server will be down briefly. This down-time is scheduled for 10:00 PM EDT / 7:00 PM PDT on July 3, 2019.

Cheers,
Gerry

 

[gerryvz]

If folks have been getting error messages of any kind over the past couple of days, that’s my fault. I upgraded the cURL software from 7.58 to 7.65 a few days back, and when compiling and installing the update, I forgot to enable SSL (essentially the capability that enables encryption). This is causing a few functions to error out or not work, such as “push” alerts in your browser.

I’m 40 minutes out from landing at Washington Dulles now, from Germany, so I’ll fix this either late tonight or more likely on Saturday morning.

No core functions of the forum are affected by this — it’s a minor annoyance/inconvenience for a few functions that the forum does automatically.

Cheers,
Gerry

 

[gerryvz]

If folks have been getting error messages of any kind over the past couple of days, that’s my fault. I upgraded the cURL software from 7.58 to 7.65 a few days back, and when compiling and installing the update, I forgot to enable SSL (essentially the capability that enables encryption). This is causing a few functions to error out or not work, such as “push” alerts in your browser.

I’m 40 minutes out from landing at Washington Dulles now, from Germany, so I’ll fix this either late tonight or more likely on Saturday morning.

I fixed the issue this morning, so we should all be good. I just reverted back to cURL version 7.58.0, and I will download and recompile version 7.65.1 with SSL support before I re-install it.

I had to do a quick reboot after reverting, so that the system would recognize the 7.58 version and OpenSSL, but the forum was only "down" for about 1 minute while doing this.

In addition, I installed about 20 system patches (Linux file updates) this morning, of which around 12 were security updates. One of the updates you'll see below is the elasticsearch search engine, which updated from version 6.8.0 to 6.8.1.

The following packages will be upgraded:

  bind9-host dnsutils elasticsearch intel-microcode libbind9-160 libdns-export1100 libdns1100 libirs160 libisc-export169 libisc169 libisccc160
  libisccfg160 liblwres160 libnss-systemd libpam-systemd libsystemd0 libudev1 systemd systemd-sysv udev

Cheers,
Gerry

 

[gerryvz]

I just a couple of minutes issued a system re-boot, to accommodate the installation of a number of core Ubuntu Linux kernel upgrades, and also an upgrade to the ImageMagick software that handles the image thumb-nailing, display and re-sizing.

The following packages will be upgraded:

linux-generic-hwe-18.04
linux-headers-generic-hwe-18.04
linux-image-generic-hwe-18.04
gvfs
gvfs-common
gvfs-daemons
gvfs-libs
imagemagick
imagemagick-6-common
imagemagick-6.q16
libmagickcore-6-arch-config
libmagickcore-6-headers 
libmagickcore-6.q16-3
libmagickcore-6.q16-3-extra
libmagickcore-6.q16-dev
libmagickwand-6-headers
libmagickwand-6.q16-3
libmagickwand-6.q16-dev 
libmagickwand-dev
snapd
xserver-common

This updated the Ubuntu Linux kernel from 4.18.0.22 to 4.18.0.24. Pretty routine weekly system maintenance.

The forum was down for approximately one minute.

Cheers,
Gerry

 

[gerryvz]

This morning I was able to do something I've been wanting to do for some weeks -- move a TON of files (about 45 GB worth!) off of the main server, to an attached "block storage" drive for ready access.

Let me go into a bit more detail about this.

• The base server that we are using for the site costs about $40.00 per month. This includes 160 GB of "built-in" storage.
• The XenForo forum software and attached images and files only use about 60-65 GB of storage.
• You can get cloud-based block storage, which can be directly attached to the server (think of it as an external hard drive on your computer) for 10 cents per Gigabyte per month. Since bringing this site up, I have had a 50GB block storage drive (which costs $5.00 per month) that has sat unused.
• I had a complete copy of the old vBulletin forum, and its ~800 MB database, on the main server for convenience, mainly doing the import of the old forum's images, database and Wiki into the new forum. All of this "old forum" data totaled about 45 GB.
• This 45GB of "old forum" data, combined with the 60-65GB of "new forum" data, totaled out around 110 GB of total data out of the 160 GB that I have available as part of the server plan (see first bullet above).
• By moving this "old forum" data to the block storage drive, I am able to store it MUCH more cheaply.
• MORE IMPORTANTLY, I am now able to reduce the size of the server plan that this forum is using, from an 8GB memory, 4 CPU cores, and 160 GB storage plan -- which costs $0.06 per hour, or $40 a month -- to a lower level plan with 4GB memory, 2 CPU cores, and 80 GB storage, for $0.03/hr or about $20 a month.
• This will dramatically reduce the monthly costs of running the site -- to the tune of a ~$250/year cost reduction.
• I did increase the size of the block storage disk from 50GB to 75GB, which will bring the total cost of the storage to $7.50 a month, in addition to the $20.00 I will be paying (down from $40).

I will be "reducing" the plan in the near future from the $40/month to the $20/month plan. The downtime for the conversion to do this SHOULD be under one hour, but the Linode folks say it would be two hours maximum.

I will give 12 hours' notice as to when I will definitively do this, but I expect it will be tomorrow (Thursday) morning around 6:00 AM Eastern Standard Time (3AM Pacific Time, Noon Central European Time). I'll confirm this definitively later today.

Cheers,
Gerry


P.S. A couple of stats: the collective attachments, PDFs and other items that comprise this forum currently total about 50 GB. The cost of running this site (after moving to the $20/month server plan) will be around 3x the total costs as with the "old" site's server, not including all of the software subscription/license/maintenance costs for this site, which total around another $150+ annually.

P.P.S. If you are interested in more info/background about "block storage" and how it works, you can find info here: https://blog.linode.com/2018/02/01/announcing-linode-block-storage-volumes/

 

[nocfn]

Gerry for OMB or Fed Reserve Chairman

 

[gerryvz]

Yesterday I installed a new add-on for the forum that is designed to do the following:

1. Track and record all links on the forum (as members click on them) that don't work, resulting in a "404 Not Found" error
2. Provide a "standard" or customizable "404 Not Found" page to display when a non-working link is clicked on

I will periodically go through the non-working (and/or non-redirecting) links that are recorded, and try to fix them. What I've already seen is that a lot of the non-working links are to photos that are no longer correctly linked, or are missing, or are references to external web sites where the photos are no longer avaialble.

The add-on has already recorded hundreds of non-working links. Per above, many if not the majority of them are non-working individual images.

In the near future I will create a cool HTML "404 Not Found" page to display, which is in keeping with the design/topic/theme of this site, to replace the generic "404" page that this add-on currently displays.

For many of the non-displaying images, I can't do anything about that. For web pages with link errors, particularly if it is somehow referring to a valid internal site or external site page, I can create appropriate re-directs or correct referrals to the intended page.

Small tweak, but something that should (over time) help improve the overall site experience.

Cheers,
Gerry

 

[gerryvz]

Yesterday I installed a new add-on for the forum that is designed to do the following:

1. Track and record all links on the forum (as members click on them) that don't work, resulting in a "404 Not Found" error
2. Provide a "standard" or customizable "404 Not Found" page to display when a non-working link is clicked on

I will periodically go through the non-working (and/or non-redirecting) links that are recorded, and try to fix them. What I've already seen is that a lot of the non-working links are to photos that are no longer correctly linked, or are missing, or are references to external web sites where the photos are no longer avaialble.

The add-on has already recorded hundreds of non-working links. Per above, many if not the majority of them are non-working individual images.

In the near future I will create a cool HTML "404 Not Found" page to display, which is in keeping with the design/topic/theme of this site, to replace the generic "404" page that this add-on currently displays.

For many of the non-displaying images, I can't do anything about that. For web pages with link errors, particularly if it is somehow referring to a valid internal site or external site page, I can create appropriate re-directs or correct referrals to the intended page.

Small tweak, but something that should (over time) help improve the overall site experience.

Cheers,
Gerry

A little tweaking and coding on my part produced a 404 ("Not Found") screen with a search box to assist the user..... bit more increased functionality.

 

[gerryvz]

I've been reviewing a lot of the 404 URLs that the site has been receiving, per the add-on that I mentioned above. Then comparing them to what others have been finding.

What's happening is that the vast majority (probably 95%) of the 404 URLs ("Not Found") are being caused by web crawlers and spiders, which had recorded URLs on the old forum, and adapted them nearly verbatim to this new forum, expecting to get redirected to an equivalent page on this forum. And this forum has no equivalent page, and the thread/post/forum redirects (which are working just fine) are not applicable to these esoteric pages that the old forum had.

 

[gerryvz]

With today’s migration to a smaller server, I also implemented a backup regimen that is offered by Linode, for $5.00 a month. This is a multi-faceted backup that provides multiple layers of daily, weekly and monthly backups (retaining separate copies for each of those periods) of all 500Eboard files on their servers. It also gives me the ability to do an instant “snapshot” backup at any time.

This is all in addition to the weekly, manual backup that I personally do of our site’s “crown jewel” — the MySQL database.

If anything ever happened or we lost the site, we now have automatic and full backups to restore from. Extra safety and security.

 

[gerryvz]

I just received the following note from the folks at Linode, our hosting provider:

Hello,

Maintenance is required for one or more of your Linodes' physical hosts. This maintenance is necessary to patch the MDS (ZombieLoad) CPU vulnerability[1] and will incur downtime for your Linode(s).

Some FAQ:

- Your maintenance schedule can be found at the bottom of this ticket. The scheduled start times are also listed within the Linode Manager[2/3].
- These maintenance windows cannot be altered, postponed, or rescheduled, and rebooting your Linode ahead of time will not remove the need for this maintenance.
- A two-hour window is allocated, however, the actual downtime should be much less.
- Your Linode(s) will be cleanly shutdown and will remain inaccessible during the maintenance window. After the maintenance has concluded, each Linode will be returned to its last state (running or powered off).
- These updates affect the underlying infrastructure that your Linode(s) reside on and will not affect the data stored within your Linode.

To fully mitigate the MDS vulnerability, set your Configuration Profile to utilize our latest kernel. If you use a distribution-supplied kernel, you can update your kernel[4] via your package manager. Linodes that use our latest kernel will automatically update upon reboot.

To ensure your Linode's essential services (MySQL, Apache, NGINX, etc.) return successfully upon reboot, we strongly recommend that you review our Reboot Survival Guide[5] prior to the start of the maintenance period. If you encounter any connectivity issues after the maintenance is complete, please refer to this Community Site post[6].

Please let us know if you have any questions or concerns. Thanks!


This means that the server will be down briefly. This down-time is scheduled for 10:00 PM EDT / 7:00 PM PDT on July 3, 2019.

Cheers,
Gerry

I want to remind everyone that this system maintenance / security patch at the infrastructure level, by the hosting provider Linode, will be happening on Wednesday of this week. I expect that the forum will likely be off-line for approximately 15-30 minutes during this patching window of 7PM PT / 10PM ET / 3AM BST. Perhaps less.

I have no control over this outage, as it is being initiated and performed by Linode to their own servers.

Cheers,
Gerry

 

[gerryvz]

For those interested, wanted to provide a quick update on a few things.

1. Linode system maintenance / security patch to server. From what I can tell, the system patch was applied last (Tuesday) night without any major disruption to the site. I was not online during the time frame when the patch was supposed to be applied, but this morning when I checked, the system was up and running just fine. The system logs did not indicate any major (or minor) service disruptions, so I am assuming that things went well. EDIT: as indicated here on Linode's status page, the security patch went well and was completed. https://status.linode.com/
2. 500Eboard.co domain. The 500Eboard.co domain has now been migrated over to Google Domains, from the old 1and1.com hosting provider that we used from the old forum. The 500Eboard.com domain was also migrated over to Google Domains a few days ago from 1and1, and it was pointed directly to 500Eboard.co. The next step in the process will be to migrate this site BACK to 500eboard.com as the PRIMARY domain, and keep 500Eboard.co as the backup/secondary domain. Ideally, what I will want is any URL that is typed in, whether it be 500Eboard.com or 500Eboard.co, will point to the exact same thread/post/forum, and the two domains will be interchangeable.
3. Final vBulletin site snapshot. Now that the final domain (500Eboard.co) has been migrated off of the 1and1.com hoster's management, I am in the process, this morning, of doing a FINAL download via SFTP (secure FTP) of all 60 GB of files from the old vBulletin site, to my hard drive. This will provide the final record of the "frozen" 500Eboard vBulletin site, in case I ever need the files again. They are also uploaded to this site's servers, on the block storage drive, for near-term access. But the downloading that I am doing today is a final snapshot of these files, for archival purposes.
4. 1&1 contract cancellation. Later today, when all of the files are successfully downloaded from the old 1and1.com server, I will contact them and cancel the account and initiate a refund of the remaining ~$80-100 that is left over from the pre-paid three-year contract I had with them, of which around 1 year is remaining. This will terminate the relationship that this forum has had with 1and1 since approximately 2011.

Generally, I was very happy with 1&1 over the years that we used them. With a couple of exception when we had extended (multiple days) outages, which were not the fault of this forum, the performance and reliability of their infrastructure was pretty good. We started nearing some of their critical resource limitations over the past couple of years, which unfortunately there was no way to get around. Not to mention the fact that vBulletin version 4 was no longer being supported by that company, and the reviews on their current version (vBulletin 5) were not that stellar as an upgrade path.

It has been a long and arduous road to both migrate this site not only to a new hosting provider (which would be a job in and of itself), but also to a totally new and different software platform, and create and configure a totally new infrastructure from scratch, and do it in a way that was the least disruptive as possible to the site's members.

We are closing the final chapter in the life of the "old" site and jumping with both feet into the future!

 

[gerryvz]

1. 1&1 contract cancellation. Later today, when all of the files are successfully downloaded from the old 1and1.com server, I will contact them and cancel the account and initiate a refund of the remaining ~$80-100 that is left over from the pre-paid three-year contract I had with them, of which around 1 year is remaining. This will terminate the relationship that this forum has had with 1and1 since approximately 2011

The cancellation process for the 1and1.com contract has been started. I first initiated the contract with them on January 24, 2010. That's 9.5 years ago !!

 

[gerryvz]

Linux update this morning resulted in a lot of php-related files getting updated, including for old (unused) version 7.2. Php update also brought the existing version 7.3.6 up to 7.3.7, which is the latest.

After updating the files, I deleted the older versions of php that were on the server, which were 7.0 and 7.2.

Server has been up and running around 11.5 days since last reboot.

libapache2-mod-php7.2 libapache2-mod-php7.3 libpcre16-3 libpcre2-8-0 libpcre3 libpcre3-dev libpcre32-3 libpcrecpp0v5 php7.2 php7.2-bcmath
  php7.2-bz2 php7.2-cli php7.2-common php7.2-curl php7.2-gd php7.2-gmp php7.2-intl php7.2-json php7.2-mbstring php7.2-mysql php7.2-opcache
  php7.2-readline php7.2-xml php7.3 php7.3-bcmath php7.3-bz2 php7.3-cli php7.3-common php7.3-curl php7.3-fpm php7.3-gd php7.3-gmp php7.3-intl
  php7.3-json php7.3-mbstring php7.3-mysql php7.3-opcache php7.3-readline php7.3-recode php7.3-tidy php7.3-xml php7.3-xmlrpc php7.3-zip

 

[gerryvz]

1. 1&1 contract cancellation. Later today, when all of the files are successfully downloaded from the old 1and1.com server, I will contact them and cancel the account and initiate a refund of the remaining ~$80-100 that is left over from the pre-paid three-year contract I had with them, of which around 1 year is remaining. This will terminate the relationship that this forum has had with 1and1 since approximately 2011.

The account/contract with 1&1 has now been formally cancelled, with remaining of the pre-paid funds to be deposited into my bank account in the next few days. This should contribute to about 3 months' operation of this site.

 

[gerryvz]

With today’s migration to a smaller server, I also implemented a backup regimen that is offered by Linode, for $5.00 a month. This is a multi-faceted backup that provides multiple layers of daily, weekly and monthly backups (retaining separate copies for each of those periods) of all 500Eboard files on their servers. It also gives me the ability to do an instant “snapshot” backup at any time.

This is all in addition to the weekly, manual backup that I personally do of our site’s “crown jewel” — the MySQL database.

If anything ever happened or we lost the site, we now have automatic and full backups to restore from. Extra safety and security.

If you're interested, here are the automatic back-ups that are done early every morning of the entire site's files on the Linode servers. I also do a "snapshot" backup (manually induced by me) every weekend, to do along with the manual backup of the database that I do each Sunday.

You can see all of the backups that have happened since I initiated the regimen at the beginning of July.

Cheers,
Gerry

 

[gerryvz]

This morning, I updated the Wiki software to a new beta version, from VaultWiki 2.1.0 Beta 1, Build 2 to version 2.1.0 Beta 2, Build 1.

It contains a lot of bug fixes, including fixes to a few bugs that I reported over the past month. The Wiki is functional and we are continuing to review and update the various pages on an ongoing basis, as some pages lost data on the import from the old forum.

 

[gerryvz]

Here is the changelog for the new versions of XenForo software released today:

For XenForo version 2.1.3 (this is the overall forum software)

• Don't log when a push notification hits a 406 error as this is a temporary rate limiting issue.
• Update the Microsoft connected account explain text to point to the new Azure app registration portal.
• When logging a visitor out, pass the Clear-Site-Data header to ensure that cached content which may not ordinarily be visible by the next visitor, or a guest, remains not visible. (Thanks @Steffen)
• Fix the TagContent.AddUser relation.
• Support a "default" value for numberbox elements that isn't overridden by the "min" value particularly important for custom field values.
• Set appropriate content type for connected_account.php output.
• Update "Review" phrase on the file health check list to "Results" to avoid potential multi-lingual ambiguity.
• If a numberbox step attribute is set to "any" the stepUp/stepDown built-in methods stop working so fallback to our own code.
• When the Import CLI command is complete, add a note about being able to run the import-finalize command.
• Ensure the login form, particularly when it errors, maintains its redirect URL.
• If setting values in the Schema Manager, set any existing length value to null.
• Rename Templater::fn function to Templater::func. Templater::fn remains but will be removed in the near future.
• Hide active/expired upgrades if their parent upgrade has been deleted in more cases.
• Update wording from EmojiOne to JoyPixels. Also update emoji artwork to JoyPixels v5.0.
• If a webhook signing signature is not entered, skip verifying the provided signature.
• Add missing 'poll' content type.
• Update to Froala 2.9.5 to fix an IE11 compatibility issue.
• For new upgrades from XF 2.0 to 2.1, ensure the trophy phrases are updated to reflect the new reactions system.
• When starting a new conversation or adding users to an existing conversation, make it more clear if there are recipient limits in place.
• Fix API doc typo.
• No longer set a default alt value when inserting an image into the editor.
• Bail out of updating user trophies earlier if a trophy has already been awarded.
• When escaping JSON strings, ensure the tab character is converted to '\t'.
• When inserting a quote into the editor, before attempting to focus the editor, scroll the container into view first.
• Ensure the lastPageLinks option is passed in to the conversation list too.
• Suppress a browser console error if the canvas operations to draw over a favicon image fail.
• Override some inconsistent Froala styling.
• Add a new config key serviceUnavailableCode (defaults to 503) which represents the response code returned when parts of the site are unavailable.
• Remove superfluous tooltip from cookie notice approve button.
• Force payment provider log messages to their max length.
• When displaying the users who reacted to content, ensure we fetch all required data up-front.
• Revert a change from XF 2.1.2 which affected the read/unread appearance of conversation messages.
• Ensure the date of birth can be changed on a user's profile until it has been fully completed before it is locked.
• Update language title Azeri to Azerbaijani.
• Display a more accurate summary of the count of active/expired upgrades on the user upgrades list.
• Prevent an error in payment.js if a Stripe iframe is missing a name attribute.
• Prevent double escaping of thread prefixes in post reports.
• Allow error and payment provider logs to be sorted by their primary key in order to have more accurate chronological sorting for logs that can happen quickly.
• Try to do a case insensitive look up for the correct content type when fetching metadata from a URL.
• Change visibility of overridden canBookmarkContent method to match parent
• Add a checkbox so that add-ons rebuilt/upgraded from archive can opt to have all their files overwritten rather than only changed files.
• When increasing the unread alerts count for a user, do not let the number exceed the maximum storage for the DB field (65,535).
• When rendering bookmark list items, use the built in content_user getter to check if a user value exists to render their avatar.
• Attempt to make post quotes by ignored users clearer.
• When logging content changes, return the applied changes correctly.
• Attempt to suppress some post delete actions if the record being deleted has already been deleted.
• Fix the automatic push notification subscription logic so you can be re-subscribed after logging out and logging back in.
• After re-enabling the editor, refresh the editor size to ensure it is at least the minimum height.
• Implement a clickable calendar icon on date inputs to show the date picker.
• Make the show password toggle checkbox display a pointer on hover.
• Add extra margin to off canvas navigation to prevent last category being unselectable on certain devices.
• Correctly maintain/migrate message visibility when copying/moving to a new thread.
• Hint index when fetching user content reactions
• Support the XAVC ftyp when detecting video files.
• Template error using 'exact' for user fields regardless of context
• Prevent markdown from being parsed inside custom BBCode tags set to stop parsing BBCode
• Attempt to upload images inserted via the editor using large data URIs.
• Ensure that the invalid notice criteria message isn't displayed if the page criteria controller action field is empty.
• Use a different pagination / filtering method on the bookmarks page to reduce possibility of duplicate items across pages.
• Make use of database transactions in a few entity delete cases where they currently might be skipped.

For XenForo Enhanced Search version 2.1.2

• Increase frequency of reindex retries to every 10 minutes and attempt to reindex up to 1000 items at a time.
• Attempt to drop the ES temp search index table (if exists) before creating a new one.

For XenForo Media Gallery version 2.1.3

• Use the storecat value when building the original source file location where appropriate.
• Change visibility of overridden canBookmarkContent method to match parent
• Remove misleading text from album creation page.
• Add some missing moderator log phrases and ensure entries logged when category changes occur.
• Implement RSS feeds of lists of media items for media index, category index and album index, similar to XFMG 1.x. RSS feed can be accessed by adding index.rss to the end of an existing URL.
• Use correct icon for custom thumbnail placeholder.
• Display the correct "noThumb" icon for media embeds.
• When displaying the transparent "noThumb" image, generate one which is the same dimensions as generated thumbnails if the thumbnail is non-square.
• Adjust size of delete button when editing an image note and workaround a sizing issue for the note field in Firefox.
• When an album is inside a category, check the visibility permissions (deleted/moderated etc.) first before category permissions.
• Fix incorrectly labelled "Manage any tags" permission.

 

[gerryvz]

After 25 days of continuous system up-time, I issued a reboot of the system at 8:40 AM EDT. This was because of an update that needed to be made (and applied) to the Linux kernel, from version 4.15.0-54-generic to version 4.15.0-55.60.

The forum was "down" for a total of about one minute.

More information about the Linux kernel upgarde is here:

4.15.0-55.60 : linux package : Ubuntu

launchpad.net

 

[gerryvz]

Incremental MySQL database update today, from 5.2.6 to 5.2.7. Update did a full database check, and all tables checked out OK. Restarted the MySQL service after update via systemctl restart mysql command in Linux system console.

System downtime, approximately three seconds.

Release notes for MySQL 5.7.27 are located here:

MySQL :: MySQL 5.7 Release Notes :: Changes in MySQL 5.7.27 (2019-07-22, General Availability)

dev.mysql.com

 

[gerryvz]

Very minor update of the elasticsearch engine today, from version 6.8.1 to 6.8.2. Then I issued a service elasticsearch restart command in the Linux terminal to switch over to the new code.

Elasticsearch 6.8.2 release notes located here:

Elasticsearch version 6.8.2 | Elasticsearch Guide [6.8] | Elastic

www.elastic.co

 

[gerryvz]

This week, Ubuntu issued a new version of its kernel software. This morning, approximately 8:35 AM EDT, and after a full 10 days of system up-time, I initiated a full system shut-down and then re-boot after installing the new version of the Linux kernel.

The new kernel, version 5.0.0-23, replaces the older version 4.18.0-24

Release notes for this Linux kernel: http://changelogs.ubuntu.com/changelogs/pool/main/l/linux/linux_5.0.0-23.24/changelog

Affected files include the following:

The following NEW packages will be installed: linux-headers-5.0.0-23 linux-headers-5.0.0-23-generic linux-image-5.0.0-23-generic linux-modules-5.0.0-23-generic linux-modules-extra-5.0.0-23-generic The following packages will be upgraded: linux-generic-hwe-18.04 linux-headers-generic-hwe-18.04 linux-image-generic-hwe-18.04

The system was down for approximately 1.5 minutes during the system shutdown, and then rebooted. I had not fully shut down the system for at least a couple of months.

 

[nocfn]

I was waiting for the Postal warning....

 

[gerryvz]

I was waiting for the Postal warning....

User warning screens will only apply for updates and upgrades to the forum software or other components that the forum is running on, such as add-ons.

Nearly all system-level upgrades and maintenance, including the Ubuntu operating system and associated kernel files, Apache web server and MySQL database, generally require a system reboot to process and initiate the new files. It makes no sense to use a "Postal" warning screen when the entire system is going to be brought down, and then back up again.

Rebooting the system will not allow the display of any warnings, because all system services that underpin the forum are stopped and then re-started. Fortunately, and unlike WinBlows or even a Mac, a full system shutdown and reboot of a Linux system is generally achieved within 60-90 seconds.

Other upgrades of system software and other components running on the system are "slipstreamed" in, meaning that the files are upgraded, and then that specific/individual service is stopped and restarted. This takes place in under five seconds, and is generally imperceptible to users.

One example of a "slip-stream" upgrade was Tuesday's upgrade of the search engine, as I detailed in the post above: FYI - Technical Site News (for those interested)

 

[gerryvz]

Starting at around 6:00 PM EDT / 3:00 PM PDT today, Aug. 8, I am going to take the site down for approximately 15 minutes.

The reason for this is so I can create a second domain (500Eboard.com) on the server, and I need to instruct the Apache web server that this new domain is there and available. This is a fundamental and important step to getting the thread/post redirections working again for URLs that reference the old 500Eboard.com threads.

Basically I'll just be creating a bunch of new folders and directories on the server, but I have to have Apache NOT RUNNING when I do this so that when I bring Apache back up, it "sees" the new 500Eboard.com domain and its associated folders and logs.

Again, will be 10-15 mins of downtime at the VERY MOST -- maybe as little as 5 mins.

Thanks
Gerry

 

[alabbasi]

15 minutes!! Oh my god, OH MY GOD , O H M Y G O D! ! ! !

 

[gerryvz]

First part of the mission accomplished. DNS records have been created, the domain and associated files have been created on the server, and Apache (and this forum) were brought down and back up again. Apache is now aware that there is a second domain on the server.

Now all I need to do is create the code on the server to tell the 500Eboard.com requests to go to 500Eboard.co. This is more difficult than a standard 301 or 302 redirect — it will be done via a carefully coded .htaccess file.

The forum’s redirect system will do the rest. Perhaps this will be up & working tomorrow — we’ll see.

 

[gerryvz]

First part of the mission accomplished. DNS records have been created, the domain and associated files have been created on the server, and Apache (and this forum) were brought down and back up again. Apache is now aware that there is a second domain on the server.

Now all I need to do is create the code on the server to tell the 500Eboard.com requests to go to 500Eboard.co. This is more difficult than a standard 301 or 302 redirect — it will be done via a carefully coded .htaccess file.

The forum’s redirect system will do the rest. Perhaps this will be up & working tomorrow — we’ll see.

Here's a screen shot of the server, showing both domains side by side, in folders. The 500Eboard.com domain folder is the new one (you can see today as being the "modification" date.

Just FYI, every single file in the forum is inside that "500Eboard.co" folder.

This is evidence that the first steps I described above were successful.




Here is a view of the directories, as seen from the terminal via secure shell.

At the bottom, you can see the two directories (folders) in purple text. I then change to the 500Eboard.com directory (the cd command) and then I list for you (the ls command) the contents of the 500Eboard.com directory.



Ready to proceed to the next steps.

 

[gerryvz]

I want to let everyone know that redirects are now working again, and this should be permanent and not broken again in the future.

The only exception to this is Wiki links, unfortunately. I'll see what I can do for this.

However, I'm very proud that the redirects once again are operational. Please let me know if you have any problems.

I was feeling very motivated this morning, after a couple of cups of Peet's Major Dickason's Blend coffee in my French press, and went on a mini-programming spree to create the redirect code. I saved this in a file, which I then uploaded to the "new" 500Eboard.com domain's public_html folder. But first, I had to create an empty "forums" file in the public_html folder, to mimic the base structure of the "new" 500eboard.co site.

All is good, and I think we are on the road to having permanent redirects now.

 

[gerryvz]

Did a "slipstream" update to a new build of the Wiki software today. It's still version 4.1.0, Beta 2, but the software moved from build 001 to build 002 with this update.

The following items were involved with this version.

4.1.0 Beta 2 ============ Fixed Bug: [i#5810] bb-codes deleted after XF 1.x upgrade [in: XF >= 2.0] Fixed Bug: missing template for pre-4.1.x edit alerts [in: XF] Fixed Bug: fatal error viewing most wiki-related alerts [in: XF >= 2.0] Fixed Bug: [i#5796] default library/vw/config.php.xf points to wrong path [in: XF < 2.0] Fixed Bug: [note#31876] fatal error getting index when upgrading from pre-4.1.0 Alpha 1 Fixed Bug: [i#5738] wrong phrase for htmlstate changes in history [lite: no] Fixed Bug: [note#31872] database error upgrading from pre-4.0.19 Fixed Bug: [note#31869] pre-4.0.18 upgrade hangs [in: XF >= 2.0] Fixed Bug: [note#31864] infinitely updating dimensions upgrading from pre-4.0.17 if exactly 1 entry affected [lite: no] Fixed Bug: [i#5794] fatal error deleting disallowed titles [lite: no] Fixed Bug: [i#5794] fatal error saving disallowed titles w/ "any" prefix [lite: no] Fixed Bug: [i#5816] some special pages show empty-text notice [lite: no] Fixed Bug: [i#5808] content wrapped in any other BB-Code is not autolinked [in: XF >= 2.0] [lite: no] Fixed Bug: some incorrect attachment dimensions after upgrading from pre-4.0.17 [lite: no] Fixed Bug: [note#31812] fatal errors upgrading from pre-4.0.16 [in: XF >= 2.0] Fixed Bug: duplicate title error for matching personal feed titles Fixed Bug: resolved conflict w/ NixFifty_AMP [in: XF <= 2.0] Fixed Bug: [i#5804] fatal errors running some 4.0.x steps upgrading from XF 1.x [in: XF >= 2.0] [lite: no] Fixed Bug: conflicting synonym created when editing index [in: XF >= 2.0] [lite: no] Fixed Bug: E_NOTICE renaming wiki content [in: XF >= 2.0] Fixed Bug: can't resolve route-filtered wiki pages at different base URL [in: XF] [lite: no] Fixed Bug: invalid route error for route-filtered wiki pages [in: XF >= 2.0] [lite: no] Fixed Bug: route filters replacing vw/ not applied [in: XF] [lite: no] Fixed Bug: feed only shows oldest entry on each page [since: 4.1.0 Alpha 1] Fixed Bug: fatal error adding content to a node on content-create Fixed Bug: route-filters for vw-feed/ route inconsistently applied [lite: no] [in: XF >= 2.0] Fixed Bug: [i#5793] fatal error in rebuild after renaming/removing prefix [lite: no] Style Tweak: [i#5794] improved admin-style consistency of Disallowed Titles [lite: no] == BUILD 001 == New Feature: [i#2316] Allow/Block Actions Per-User Per-Content [lite: no] New Feature: Support for XenForo Spam-Cleaning [in: XF >= 2.0] New Feature: [i#4768] Option for Maximum Visible TOC Depth [lite: no] Changed: [i#5740] simplified context of tabs above new page editor Changed: [i#5754] separate admin version of upgrade-pending message in public areas [in: XF >= 2.0] Fixed Bug: [i#5791] can't create new bb-codes if any prefix has no tag [in: XF >= 2.0] [lite: no] Fixed Bug: [i#5784] other class extensions to \XF\Mvc\Router are overridden [in: XF >= 2.0] [since: 4.1.0 Beta 1] Fixed Bug: undefined variable in task when re-enabling VaultWiki [in: XF >= 2.0] [since: 4.1.0 Beta 1] Fixed Bug: raw code in output of template using PLAIN on parameter [in: XF >= 2.0] [lite: no] Fixed Bug: fake style-properties not merged if ::get before template render [in: XF >= 2.0] Fixed Bug: fatal error sending wiki push notifications while wiki in maintenance mode [in: XF >= 2.1] Fixed Bug: database error when multiple users access same content w/ unprimed autolink cache at same time [in: XF >= 2.0] [lite: no] Fixed Bug: invalid arguments in handlers while wiki not active [in: XF >= 2.0] Fixed Bug: data not saved if query after some failed DMs [in: XF >= 2.0] Fixed Bug: VW4 import: fatal error saving comment attachment import IDs [lite: no] [since: 4.1.0 Alpha 1] Fixed Bug: VW4 import: fatal error importing comment attachments [lite: no] [in: XF >= 2.0] Fixed Bug: VW4 import: fatal error importing reports [lite: no] [since: 4.1.0 Alpha 1] Fixed Bug: VW4 import: database error importing integrations if forum wasn't an import [lite: no] [in: XF] [since: 4.1.0 Alpha 1] Fixed Bug: VW4 import: fatal error importing team-managed pages [lite: no] [since: 4.1.0 Alpha 1] Fixed Bug: VW4 import: content tags for non-pages not imported [lite: no] [since: 4.0.16] Fixed Bug: VW4 import: fatal error importing content tags [lite: no] [since: 4.1.0 Alpha 1] Fixed Bug: fatal error parsing BB-Code of non-existent content [in: XF >= 2.0] Fixed Bug: minified cp-home.js calls hasClass on null Fixed Bug: invalid import config directory [in: XF >= 2.0] [lite: no] Fixed Bug: exceptions not logged if debug mode disabled [in: XF >= 2.0] Fixed Bug: [note#31692] empty parameters treated as non-empty by IF [in: XF >= 2.0] [lite: no] Fixed Bug: index WIKI links render like optionless links Fixed Bug: VAR local variables not working [lite: no] [since: 4.1.0 Alpha 1] Fixed Bug: upper book TOC entries after nest appear in nest [lite: no] Fixed Bug: book TOC levels not reaching max depth for some values [lite: no] Fixed Bug: invalid Show All link in book preview [in: XF >= 2.0] [lite: no] Fixed Bug: [i#5768] missing vw_misc permissions [in: XF >= 2.0] Fixed Bug: invalid URL for creating new help entry [in: XF] Fixed Bug: [i#5754] wrong upgrade-pending message in admin areas [in: XF >= 2.0] Fixed Bug: [i#5767] wrong URLs for Wiki Active status links [in: XF >= 2.0] Fixed Bug: inline-mod controls for comments appear twice [in: XF >= 2.0] Fixed Bug: selected comment not passed to inline-mod [in: XF >= 2.0] Fixed Bug: changes to feed-tab option don't clear nav-tab cache Style Tweak: new synonym prefix/title fields on separate lines [in: XF < 2.0] [lite: no] Style Tweak: [i#5749] refer to protected titles as disallowed [in: XF >= 2.0] [lite: no] Style Tweak: [i#5712] search buttons inconsistent w/ buttonGroup CSS [in: XF >= 2.0] Style Tweak: new-discussion button inside discussion-list border [in: XF >= 2.0] [lite: no] Style Tweak: no padding around no-comments text for empty discussion [in: XF >= 2.0] Style Tweak: no padding between placeholder text and categories [lite: no] Style Tweak: [i#5722] possible empty browsing-users contents Style Tweak: [i#5754] iconic pending-upgrade messages [in: XF >= 2.0] Style Tweak: [i#5747] separate radio/checkbox explanation handler Style Tweak: [i#5728] page tabs should be horizontally scrollable [in: XF >= 2.0] Style Tweak: [i#5767] use rowMessage styles for Wiki Active statuses [in: XF >= 2.0] Style Tweak: [i#5742] label misalignment for some inputs [in: XF >= 2.0] Style Tweak: [i#5723] too much padding between comments [in: XF >= 2.0] Style Tweak: no styling for category column lists [in: XF >= 2.0] [lite: no]

 

[gerryvz]

I just issued a system reboot, as an updated Linux system kernel was installed. Upgrade was from 5.0.0-23 to 5.0.0-25.

 

[TerryA]

?? It’s all Greek to me.

Thanks anyway for doing what you do

 

[gerryvz]

?? It’s all Greek to me.

Thanks anyway for doing what you do

A forum takes regular / constant care and feeding. It's not just moderating posts and such (copying, moving, correcting, editing, deleting spam, and so forth), but also regular monitoring of the back end. I do this on a daily basis. I have a lot of real-time graphs and information that monitor the overall health of the computer that runs this site.

On a couple of occasions since mid-March, when I debuted the new system, I have had "rogue" processes happen where something went haywire (for example, the search system ran out of control and started hogging 180% of the computer's processing power). So this has to be stopped, and then restarted. Several times a week, there are new versions of the software issued, and I have to make the decision as to whether I want to update it or not.

I have learned a ton about system administration with the new forum. On the old forum, all of this "back-end" stuff was done by the hosting company. The trade-off was that I had little control of anything outside of the actual forum software and the database. Now with the new forum, I do all of it, and I have 100% control over everything. It's much easier to screw things up. MUCH easier.