Come together with the global Drupal community in Rotterdam, 28 Sept – 1 Oct 2026. Sessions, contribution, connection, and Early Bird savings until 8 June.
By Sam308 on
I have multiple Drupal 4.6x, 4.7x, and 6.x sites.
All three (4) my 6.x site are getting the cron errors messages "MySQL server has gone away query" and "Cron has been running for more than an hour and is most likely stuck." I do not have any problems with the 4.6x and 4.7x sites.
Two of the 6.x sites are aggregator sites, one using the default aggregator and one using the FeedAPI aggregator.
I just installed my fifth new 6.6 site without any content and I can't access any of the administration pages such as admin/build/modules and admin/build/block because I keep getting the "MySQL server has gone away query" message.
My web hosting company, site5.com, has restarted the MySQL many times, but this is still occurring. I do not have access to the PHP or MySQL software installed on the web server.
Anybody else having these problem with Drupal 6.x?
Is this a problem with Drupal 6.x?
Sam
Comments
Yes.
I've had this error when trying to access some of my admin pages, right in the middle of site modification, in the middle of the night when access and traffic should all be fine, and for no apparent reason. Apparently, it's not unusual. Some people claim it's a problem with the server. I had to repair my watchdog table. It worked for a few days, then happened again. I repaired the table again, and it errored within minutes.
When you say, "I do not have access to the PHP or MySQL software installed on the web server."
Are you sure? You don't have phpMyAdmin in your Cpanel? You can go into phpMyAdmin, into your drupal db and run CHECK TABLE on whatever table is erroring out on you (you didn't list enough error to tell specifically). If the table comes back listing problems, you just run REPAIR TABLE on the particular table.
This is only one solution. And it only worked part-time for me.
There's a lot of discussion out there on this. See:
http://forum.siteground.com/showthread.php?t=8747
I haven't been able to solve my own particular problem, because I have much bigger ones right now. But there are lots of suggestions on that page that seemed to help some people. Good luck, man.
Partial fix... Still trying to find out the other cron problems
BTW, I do have access to the phpMyAdmin in my Cpanel. I meant no access to change the server software configurations for PHP or MySQL.
Fixed the new installation site
Anyways, as far as the fifth new installation site is concerned, I was able to fix the problem and now able to access the administration pages without the "MySQL server has gone away query" error messages.
Here is what I did.
(1) Via phpMyAdmin, I cleared the watchdog table.
(2) Went back to the site and tried to access an administration page and got the "MySQL server has gone away query" error messages.
(3) Via phpMyAdmin, I looked at the watchdog table and noticed the following message: "Unable to fetch any information about available new releases and updates.
(4) Went back to the site and modules page and disabled the Update status.
After disabling the Update status, all the "MySQL server has gone away query" error messages went away.
It appears that the Drupal Update status was causing the problems.
The Remaining four 6.x websites
I still need to find out what is causing the cron errors messages "MySQL server has gone away query" and "Cron has been running for more than an hour and is most likely stuck." for the other four 6.x sites. If I find out what the problem is, I will post the results here.
Sam
=-=
The cron error discussions that have taken place in past = http://drupal.org/search/node/Cron+has+been+running+for+more+than+an+hou...
Thanks, but....
Thanks, but I already read them.
Those previous posts are a mix of all different drupal versions over the years, so it makes it difficult to determine what is actually wrong or what is causing the problems in a Drupal 6.x version.
That is why I created this 6.x page.
Sam
=-=
All that I've read point to the same problem regardless of version and its to maniully remove the variables from the DB to stop cron from running.
I believe the same fix would is necessary for Drupal 6
I'd test using a backup of the database
RE: Remove the variables
At this point all the 6.x site cron without errors.
Were there any particular Database variables / tables you were referring to that apply to fixing the 6.x problems?
Thanks,
Sam
By the way, would you have any comments regarding my 6.x site is a FeedAPI aggegator site question in my last comment below?
Do you think this is too much for a Drupal site and will eventually choke or stall out the system?
Update on the The Remaining four 6.x websites
After disabling the Update status on all the remaining four 6.x sites, the cron seems to be working fine.
At this time (this comment post), cron has only ran one time for remaining four 6.x websites since I disabled the Update status.
If I get anymore cron errors, I will report them back here on this page.
Sam
One question (kind of unrelated):
One of my 6.x site is a FeedAPI aggegator site. It has 1403 feeds brought in via the FeedAPI module. The newly created nodes are associated with over 800 Taxonomy terms. Currently, the database is about 500 MB in size.
The website is called "Non Stop Events" at http://nonstopevents.com
Do you think this is too much for a Drupal site and will eventually choke or stall out the system?
Sam
Site5 has some pedantic
Site5 has some pedantic firewall settings and updates.drupal.org gets blocked (too many requests?) which caused the above (OP's) error message a bunch of times for me over the last month. To test, ssh to your Site5 server:
lynx to http://updates.drupal.org/ will hang for many minutes or longer if the above is the case, I solved this by opening a support ticket and suggesting the white-listing of updates.drupal.org IP (140.211.166.6 -- subject to change??)
I'm told this may reoccur when firewall is updated, requiring another ticket but for now Update Status is all better -- no more errors :)
This was a ball-tearer to find and I spent the best part of a day finding it.
I hope this helps
I'm also on Site5
I am also on Site5 and had the same MySQL-server-has-gone-away issue when running cron job and Available Updates. After submitting a support ticket to request to access updates.drupal.org the issue was resolved. Nothing was done on MySQL settings.
I like Site5 and their support reacts quickly and professionally. I just don't understand why they block updates.drupal.org.
Seeing this on Hostgator (VPS)
Seeing this on our Hostgator VPS after a server migration. If I disable the Upate Status module, the problem goes away. Will post back if that turns out ot be the long term cure in this case.
EDIT:
updates.drupal.org does not appear to be blocked; however, disabling the updates module does resolve the issue, and is the only thing so far that's helped. Mystery.
Changing providers fixed it for me
wrt OP's original post
One person in the previous link I gave suggested it had to be the provider. I found their answer to be somewhat dubious and arrogant in its tone and lack of information. However, my MySQL errors of this sort completely disappeared when I changed provider (for unrelated reasons). I did clear the watchdog db completely as opposed to repairing it, after I moved. But the problem did seem to be an _access_ problem.
I never got any cron errors and did not try the above-mentioned disabling of update status. Is disabling update status a good idea for a healthy site?
MyIsam?
Hi,
You can check this thread http://drupal.org/node/311180