Recent content by TUG

just a final update. xenforos (and my hosts) suggestion was to install cloudflare. in the meantime, host was able to block all traffic from vietnam, brazil, ukraine and russia. and that alone has wiped out 90% of the bot traffic we were seeing (sofar at least). they also modified the settings...

looking at the log file. the errors started at 3:03am in the xenforo server log indicating the table was full. they stopped on their own at 4:11am so im guessing the job runs once an hour perhaps and it just filled up nearly instantly after the 3am run time?

noticed it didnt even take 7000 "guests" to break it this evening at 4am. i suppose the regular chron job ran to clear out the table on its own as it started working about 10min ago. looking at the IPs a huge % of them all coming from 177. and 179. and other similar subnets showing from...

i guess thats the only option though ive confirmed that the server itself isnt even breaking a sweat even with the 20,000 connections so its not so much a hardware issue vs whatever limitations this particular table has that allows it to fill up and bring down the entire forum.

sadly after hours of fixing this earlier today, the situation has repeated itself starting at 4am with 20,000 connected users and the site is no longer displaying any pages vs a generic "an error occured while the page was being generated, please try again later" with the server log just showing...

had to restart mariadb in recovery mode just to access it, was able to truncate the search table..and now i am able to run the repair via the admin console (and see it actively rebuilding)

repair command failed via phpmyadmin

got the same error clicking to delete the table before recreating it. now clicking on any page on the forum generates this error: XF\Db\Exception: MySQL fetch error [2014]: Commands out of sync; you can't run this command now in src/XF/Db/AbstractStatement.php at line 230

did that, it ran for about a minute then gave me a server timeout error. I did NOT choose to delete the table before rebuilding, and yes its very.

same here, im seeing anywhere from 10 to 20 thousand, however the server itself is not anywhere near its limits in disk/ram/network/etc... my issues have all been related to these two tables with now the search table being corrupted...not quite sure how to fix that.

attempted the fixes for the activity table and while that originally got the server back up and running, now the latest round of 11-15k users generates this error: xf_search_index' is marked as crashed and last (automatic?) repair failed in src/XF/Db/AbstractStatement.php at line 230 support...

same thing happening here, cleared the activity table and restarted mariadb and forum was normal.... till 15 thousand more connections happened and now it appears the search table has crashed? site generates this error for any attempt to post/edit Table './tugbbsc_xenforo/xf_search_index' is...

support seemed to fix it by clearing out the table, and restarting mariadb, however its now crashed again but this time with this error " Table './tugbbsc_xenforo/xf_search_index' is marked as crashed and last (automatic?) repair failed"

had this happen today actually... 15k users and counting!

outstanding, thank you! sorry for being dense...been out of IT for quite some time and now feel like my parents must have when they asked me for help 20 years ago!