• Hi Guest !

    Welcome to the 500Eboard forum.

    Since its founding in late 2008, 500Eboard has become the leading resource on the Internet for all things related to the Mercedes-Benz 500E and E500. In recent years, we have also expanded to include the 400E and E420 models, which are directly related to the 500E/E500.

    We invite you to browse and take advantage of the information and resources here on the site. If you find helpful information, please register for full membership, and you'll find even more resources available. Feel free to ask questions, and make liberal use of the "Search" function to find answers.

    We hope you will become an active contributor to the community!

    Sincerely,
    500Eboard Management

BUG REPORTS DISCUSSION -- for new forum

I've been using both my iPhone and iPad pretty regularly with the site in recent days and weeks, and haven't had any problems with things loading. Nor with my Mac Pro desktop or MacBook Pro laptop. Unfortunately I don't have a PC work computer anymore (traded it in for a new MacBook Pro per company policy) so can't tell anything on that platform. I am using FiOS broadband as home home connection.

I haven't seen this as an issue, but if more people report problems, I will definitely add it to the "Bug List" and begin an investigation to see if it is anything on the server side.

Cheers,
Gerry
Thx Gerry!
 
It’s an iOS browser issue and how it handles images displayed via Xenforo’s “Lightbox” function.

Many many reports are confirming this for both iPhones and iPads. Android based mobile devices are working fine — no negative reports.

Reloading the image got the whole thing. Never had it happen on a desktop browser. Ever.

I recommend switching either to an Android device, or desktop browser.
 

Attachments

  • 9CC48EF2-29E4-4276-B14F-E0C80623452F.png
    9CC48EF2-29E4-4276-B14F-E0C80623452F.png
    2 MB · Views: 2
  • 4F687D2A-527C-4390-8BA8-8AD5FBB39782.png
    4F687D2A-527C-4390-8BA8-8AD5FBB39782.png
    4.2 MB · Views: 2
It’s an iOS browser issue and how it handles images displayed via Xenforo’s “Lightbox” function.

Many many reports are confirming this for both iPhones and iPads. Android based mobile devices are working fine — no negative reports.

Reloading the image got the whole thing. Never had it happen on a desktop browser. Ever.

I recommend switching either to an Android device, or desktop browser.

Odd and unfortunate! I love this forum, but not enough to dump my iphone. I’ll be soldering on with the desktop and partially loaded images on my iOS until whatever nitty OS bug rights itself. Perhaps never!

Thanks for the research.
 
XenForo seems to be considering switching to a new third-party lightbox subroutine in the future, to help combat this issue. It is not a high priority for them given that it only affects iOS devices and not other mobile or desktop devices.
 
XenForo seems to be considering switching to a new third-party lightbox subroutine in the future, to help combat this issue. It is not a high priority for them given that it only affects iOS devices and not other mobile or desktop devices.
Weird. Probably a good thing nobody uses iOS devices.

*hugs Windows XP desktop workstation*

:watchdrama:
 
FYI-
I accidentally punched in the old forum URL and was greeted with this warning.

View attachment 86816
I'm aware, thanks. It's a small issue -- basically the SSL certificate that I have for the .co site (the little green padlock in your browser URL bar) only applies to that domain, not the .com domain that I just "parked" next to the .co domain on the server. I just have to get an SSL certificate for the .com domain, which I'll do today. This won't be an issue for much longer.
 
FYI-
I accidentally punched in the old forum URL and was greeted with this warning.

View attachment 86816
I'm aware, thanks. It's a small issue -- basically the SSL certificate that I have for the .co site (the little green padlock in your browser URL bar) only applies to that domain, not the .com domain that I just "parked" next to the .co domain on the server. I just have to get an SSL certificate for the .com domain, which I'll do today. This won't be an issue for much longer.

Hi @ace10,

I obtained a new SSL certificate for the "new" 500Eboard.com domain on the server, and everything is working correctly at this point.

I ran two tests of the SSL capability, for the raw 500Eboard.com and www.500Eboard.com URLs, and both came back and tested fine. They are being served by the Apache server securely, and redirecting appropriately to the 500Eboard.co domain.

You can see the results of these tests below.
Screen Shot 2019-08-09 at 11.32.21 AM.jpg

Screen Shot 2019-08-09 at 11.38.01 AM.jpg

You can see the tests covering this site's IPv4 (XX.XXX.XXX.XX) address and IPv6 (2600:XXXX:X:X:XXXX:XXXX:XXXX:XXXX) addresses. IPv6 is very important for the future of the internet, as we have essentially run out of IPv4 addresses. More info about ipv6 here, if you are interested: IPv6 - Wikipedia

Below, I also show the work (logged into the site's server via secure terminal shell) that I did in the console to obtain and apply the SSL certificate to the 500Eboard.com domain, starting with a "date" command to show you what the actual time and date I started the process was.

Screen Shot 2019-08-09 at 11.27.32 AM.jpg

One thing needs to be done now, and that is to do a basic redirect of the 500Eboard.com domain to 500Eboard.co/forums. This will avoid the "403 Forbidden" message that now appears when you type in 500Eboard.com, and sends the user to the forum's main page. This takes about 3 minutes to create, and I should be able to get to it a bit later today.

The main thing now, is when you DO type "www.500eboard.com" you see the green SSL encryption padlock in your URL bar, indicating that things are being served securely.

Hope this helps,
Cheers,
Gerry
 
Last edited:
One thing needs to be done now, and that is to do a basic redirect of the 500Eboard.com domain to 500Eboard.co/forums. This will avoid the "403 Forbidden" message that now appears when you type in 500Eboard.com, and sends the user to the forum's main page. This takes about 3 minutes to create, and I should be able to get to it a bit later today.
:update:

I have now successfully created this redirect.

When you type (or click) on www.500Eboard.com you will automatically be redirected to the main page of this forum.

When you type (or click) a specific URL from the old forum (such as this for example: https://www.500eboard.com/forums/showthread.php?t=13003), you will automatically be redirected to that same thread here on the new forum.

I now declare the redirect problem .... SOLVED for good. :wahoo:
This will be good for our Google search results, over time as Google catches up with the redirected threads. It will take some years to do this fully.

Cheers,
Gerry
 
A new release of the Wiki software was recently released, just before the developer went on medical leave yesterday for the next month or so.

We have been on Beta 2 Build 2 for the past few weeks, but there have been a couple of bugs (which most of you can't see) that generate some system errors in the forum's administrative console when certain operations are performed.

The new version (Beta 3 Build 1) fixes these bugs, but unfortunately contains some errors that don't allow the upgrade to be properly performed. There is a workaround (modifying a few of the Wiki program files) that I will likely be doing in the next day or so, so that I can properly upgrade the Wiki software.

If you don't see the Wiki in the main (top) menu bar, just know that I have temporarily disabled it to perform the upgrade. If you DO see "WIki" in the menu bar, then all is well and continue to use it as you need to.

Cheers,
Gerry
 
We have had an issue with search that requires the rebooting of the search engine (not the entire site) about once a week. This has been a niggling problem since I brought the search online back in June, 2019. For some reason, the Elasticsearch engine stops communicating with the rest of the site, but usually a reboot takes care of it and resets things. It seems that when the search count gets up to around 250,000 searches (each search that a member makes counts for around 2-10 searches as part of this number), then things go haywire, and the search engine fails. I believe it is because it is somehow running out of memory, likely due to a memory leak either in Elasticsearch or Java, that is well beyond my own control.

Since Elasticsearch upgraded their software from 7.10.x to 7.11 about a week ago, they made some un-documented changes that have appeared to bork the search function much more frequently than before ... I have been rebooting search about once per day in the past few days. Other XenForo forum owners who are using a similar search setup, have also noticed this problem.

I'm having some discussion with the XenForo developers about what we can do about this. It seems that it is not really their problem, and that it may rather be an issue with Elasticsearch.

If things get too bad, I have another search engine setup that I may employ. @xfadmin told me that he also is planning to use this other search engine setup with the 126board.com site, interestingly.

When I log into the forum's admin console, I see the errors immediately. Usually I log in at least a couple of times a day, so catch a borked search function within a few hours of anything happening. In this case, I didn't log in since yesterday afternoon, and didn't see it until about a half-hour ago.
 
I'm going to look at the system logs today, from the previous failures of the search, to see what is happening. I have a sneaking suspicion that it is running out of memory, and then the Elasticsearch server shuts down and does not communicate with XenForo, resulting in the errors. Monitoring everything now since I rebooted this morning, it is working fine and everything is communicating as it should.

I'll continue to monitor today, but will look at the system logs and report back.

:update:

I made a couple of changes to the config file for Elasticsearch. We'll see if that makes a difference.

Restarting Elasticsearch again, and checking status, shows everything is normal, as of now.
Screen Shot 2021-02-12 at 11.17.28 AM.jpg


and then more status checking for connectivity between the servers:

Screen Shot 2021-02-12 at 11.20.14 AM.jpg
 
Last edited:
Things continued to fail with the Enhanced Search capability throughout the day yesterday, unfortunately.

More research resulted in new knowledge. The Elasticsearch version that was just released (7.11) changed how it reads and acknlowledges preferences, from all previous versions. This meant that I had to create a new Elasticsearch configuration preference file, and put it in a new location where it is readable. I also had to explicitly specify the maximum heap memory size of 2GB for the search capability (the same as before) in the new file/location.

I then restarted the search engine so that it would read and pick up the new configuration preferences file. It's been a couple of hours and all has been good ..... SO FAR. We'll see in a few more hours. Not sure whether this is a true fix or not.

Other forum owners have noted the same problem as we have had.

If things get bad enough, I'll contact the Elasticsearch folks and see what I can do.
 
FYI. A search for that name does not find the thread with that name unless a wildcard is added to the end. (I didn't paste it here, because then the search result was showing THIS post.)


1613491053880.png



Thread / post #30:

1613490833473.png
 

Attachments

  • 1613490908307.png
    1613490908307.png
    11.5 KB · Views: 1
Zero errors and the Elasticsearch has been up and running just fine since this morning -- no errors, no need for a reboot. So far.
We have been continuing to have problems with the search function crashing every 3-6 hours. I have not had time to look at the elastic search system logs to pinpoint the error.

in the meantime, I have created a Band-Aid: a “cron” job that restarts the elastic search server automatically every 4 hours. A cron job is an automation system that allows the sysadmin to automate tasks. The daily site database backups are done via cron job, for example.

this automated restarting should keep the elastic search running even if it goes down, and restart it if it fails.

lots more work to do on this over the weekend.
 
I am happy to report that my "Band-Aid" cron job is working just fine, and keeping the elasticsearch engine up and running with no issues, and throwing no errors. This is a good thing, so I don't need to disable the search engine, thankfully. It does a good job, so I REALLY wouldn't want to have to do this.

All that said, I have to delve into what the problem is with the periodic lack of communication between the search engine and the web server. This is an issue that we've had since I first implemented it, but it would only pop up every week or two in the past. Until I installed the latest version of the search engine software, which make some under the hood changes that evidently messed things up.

Nothing definitive as of yet, but I've put a few hours over the past few days into trying to research and solve the problem. It may require a complete new installation of the search engine setup, plus Java that it runs on, and so forth.
 
Still borked today.

XFES\Elasticsearch\ConnectException: cURL error 7: Failed to connect to 127.0.0.1 port 9200: Connection refused (see libcurl - Error Codes) in src/addons/XFES/Elasticsearch/Api.php at line 391
  1. XFES\Elasticsearch\Api->request() in src/addons/XFES/Elasticsearch/Api.php at line 90
  2. XFES\Elasticsearch\Api->version() in src/addons/XFES/Elasticsearch/Api.php at line 107
  3. XFES\Elasticsearch\Api->majorVersion() in src/addons/XFES/Search/Source/Elasticsearch.php at line 381
  4. XFES\Search\Source\Elasticsearch->getCommonSearchDsl() in src/addons/XFES/Search/Source/Elasticsearch.php at line 161
  5. XFES\Search\Source\Elasticsearch->getKeywordSearchDsl() in src/addons/SV/SearchImprovements/XFES/Search/Source/Elasticsearch.php at line 125
  6. SV\SearchImprovements\XFES\Search\Source\Elasticsearch->getKeywordSearchDsl() in src/addons/XFES/Search/Source/Elasticsearch.php at line 145
  7. XFES\Search\Source\Elasticsearch->search() in src/XF/Search/Search.php at line 183
  8. XF\Search\Search->XF\Search\{closure}() in src/XF/Search/Search.php at line 212
  9. XF\Search\Search->executeSearch() in src/XF/Search/Search.php at line 179
  10. XF\Search\Search->search() in src/XF/Repository/Search.php at line 34
  11. XF\Repository\Search->runSearch() in src/XF/Pub/Controller/Search.php at line 430
  12. XF\Pub\Controller\Search->runSearch() in src/XF/Pub/Controller/Search.php at line 117
  13. XF\Pub\Controller\Search->actionSearch() in src/addons/SV/SearchImprovements/XF/Pub/Controller/Search.php at line 58
  14. SV\SearchImprovements\XF\Pub\Controller\Search->actionSearch() in src/XF/Mvc/Dispatcher.php at line 350
  15. XF\Mvc\Dispatcher->dispatchClass() in src/XF/Mvc/Dispatcher.php at line 257
  16. XF\Mvc\Dispatcher->dispatchFromMatch() in src/XF/Mvc/Dispatcher.php at line 113
  17. XF\Mvc\Dispatcher->dispatchLoop() in src/XF/Mvc/Dispatcher.php at line 55
  18. XF\Mvc\Dispatcher->run() in src/XF/App.php at line 2326
  19. XF\App->run() in src/XF.php at line 488
  20. XF::runApp() in index.php at line 20
 

Who has viewed this thread (Total: 1) View details

Who has watched this thread (Total: 2) View details

Back
Top