Is anyone else having issues with the database or is it just me?
I’m trying to upload my activator log & S2S chases from earlier today & it won’t let me select the association.
Am I being stupid or is there a problem with it?
Just gave it a try.
Dropdown menu is empty indeed.
On
https://www.sotadata.org.uk/en/upload/activator
And the status page is also gone for good
UNABLE TO POST CHASER INFO, AT 1717Z JUST NOW.
The database is under a high load. I’ve restarted it and hopefully things will settle down. We are monitoring the situation. Thanks.
1 Like
Going to revert to older advice while problem exists. Will post chases after EU goes to sleep, and get word that database is working again.
Spots are working, but SOTAData is still having trouble. Hang tight, we will have it sorted soon.
It is down here for the moment.
Jan
Yep, getting “DBPROCESS is dead or not enabled”
SQL Server has gone away … i suppose it is the holiday season 
I have the same issue:
An error occurred while uploading your file. Please check the file for correctness.
Server message:
Fatal error: Uncaught PDOException: SQLSTATE[HY000]: General error: 20047 DBPROCESS is dead or not enabled [20047] (severity 1) [select AssociationID from Associations where AssociationCode=‘SP’] in /var/www/include/utils.inc:1400 Stack trace: #0 /var/www/include/utils.inc(1400): PDOStatement->execute() #1 /var/www/api/admin/secure/verify_csv(105): summitIDFromCode(Object(PDO), ‘SP/SZ-007’) #2 {main} thrown in /var/www/include/utils.inc on line 1400
Guys,
IT’S BROKEN
Poking it won’t help till we find out what is not working.
2 Likes
As I said above:
> Spots are working, but SOTAData is still having trouble.
We will have this resolved as soon as we can. We are working on it now.
2 Likes
Sorry if I upset anyone … I know this is all run by volunteers and I should be more understanding
Unfortunately the last 20 years of my career were in app development and support shouting “where in hell has the database gone ?” to the DBAs
Old habits die hard
Your efforts are appreciated
Regards
Rick
5 Likes
o
I think thanks is in order to the SOTA management & support team who are working on this behind the scenes in an attempt to get it fixed for us. As volunteers, I’m pretty sure this wasn’t how they wanted (or expected) to be spending their evening.
I’m guessing it’s like most situations in life in so much that people only notice when things go wrong, usually resulting in a barrage of complaints & aggravation. Unfortunately nobody ever notices or says anything when it goes right!
Thanks for your efforts.
5 Likes
The problem is Andrew VK3ARR has produced a set of software components that works rather well and is so stable that when it goes wrong it takes a while for the rest of us to figure out which bit is broken.
It’s very easy to turn everything off and on again and hope. But doing that often looses valuable data about what failed.
There were multiple failures tonight starting with the database SQL engine using all the memory and then stopping. I’m sure it’s meant to restart in those occasions but I think that when it stopped responding, the software talking to it also got very upset and broke.
We restarted the database SQL engine and there was still a big issue. Myself and Josh (Josh is at ‘work’ now and I am at home) spent the next 3 hours working out various scenarios to locate what we thought was fubar’d and re-starting processes or entire servers. Andrew’s design is such that we could individually reboot the 2 servers that handle the SOTAwatch requests, one at a time without taking SOTAwatch offline and most of you wont have even noticed that SOTAwatch was not so snappy in responding for a few minutes. The SOTA cluster never noticed the sequential reboots nor did the SMS server or RBNhole.
We finally homed in on the faulty component and restarted its process and have restored database operation. Everything should now be back to normal.
All I can say is I’m sorry it took so long to sort this after it was reported. But it meant I didn’t spend 2hrs doing the association updates tonight but 3hrs back and forth with Josh wondering why we were being a bit inept with a simple problem to fix.
I’m off for a very large G&T now.
10 Likes
Thanks James. Everything should be working again. Thanks to @NR7Y who alerted me to the problem via Slack within a few mins of the outage. The spotting came right back up and there was only a few minutes of outage. SOTAData took a bit longer due to reasons Andy outlined above.
73
3 Likes
Sounds like you deserve it!
Thanks again for your efforts.
Josh issued the command that brought it back to life but he can’t have a drink for a few more hours yet, he’s the man who needs thanking most.
7 Likes