rpi_logo
MW@H DBase problems
MW@H DBase problems
log in

Advanced search

Message boards : Number crunching : MW@H DBase problems

Author Message
Profile Cliff
Avatar
Send message
Joined: 28 Nov 14
Posts: 51
Credit: 83,278,045
RAC: 138,326

Message 67036 - Posted: 7 Feb 2018, 17:25:43 UTC

Lst night once again there was the unable to open dbase error, so no results could be uploaded.
Can whomever is in charge of feeding the dbase hamsters ensure that they DO get their nosh?
This dbase problem has been ongoing for some time [years in fact] Them poor 'amsters must be starving...
____________
Regards,
Cliff.
--
Been there Done That, still no Damn T-Shirt

Profile mikey
Avatar
Send message
Joined: 8 May 09
Posts: 2183
Credit: 232,361,889
RAC: 230,124

Message 67038 - Posted: 8 Feb 2018, 12:04:55 UTC - in response to Message 67036.

Lst night once again there was the unable to open dbase error, so no results could be uploaded.
Can whomever is in charge of feeding the dbase hamsters ensure that they DO get their nosh?
This dbase problem has been ongoing for some time [years in fact] Them poor 'amsters must be starving...


Seems to happen alot doesn't it? I would love to bring more pc's here but am worried about the dbase problems too, I have people to pass that I just can't.
____________

Profile Cliff
Avatar
Send message
Joined: 28 Nov 14
Posts: 51
Credit: 83,278,045
RAC: 138,326

Message 67048 - Posted: 9 Feb 2018, 23:20:17 UTC - in response to Message 67038.

Lst night once again there was the unable to open dbase error, so no results could be uploaded.
Can whomever is in charge of feeding the dbase hamsters ensure that they DO get their nosh?
This dbase problem has been ongoing for some time [years in fact] Them poor 'amsters must be starving...


Seems to happen alot doesn't it? I would love to bring more pc's here but am worried about the dbase problems too, I have people to pass that I just can't.[/quote]

Yeah, its becoming rather too much of a routine event. I've taken one of my rigs off MW@H so far, but the project is a good one, so the rest are still working MW@H.
____________
Regards,
Cliff.
--
Been there Done That, still no Damn T-Shirt

Profile Keith Myers
Avatar
Send message
Joined: 24 Jan 11
Posts: 149
Credit: 102,237,241
RAC: 20,867

Message 67049 - Posted: 10 Feb 2018, 2:03:30 UTC

I guess my utilization is low enough that I have never noticed the problem. I process MW tasks every day but only about 30 a day, so have never noticed an inability to upload because MW is such a small percentage compared to Seti.
____________

Profile mikey
Avatar
Send message
Joined: 8 May 09
Posts: 2183
Credit: 232,361,889
RAC: 230,124

Message 67052 - Posted: 10 Feb 2018, 11:51:25 UTC - in response to Message 67049.

I guess my utilization is low enough that I have never noticed the problem. I process MW tasks every day but only about 30 a day, so have never noticed an inability to upload because MW is such a small percentage compared to Seti.


I've got 3 pc's here right now but am about to another goal for me at Einstein and PG is not far behind it, I'm 3rd on my Team here at MW but the 2 people ahead of me stopped crunching long ago. I would love to pass them but just can't bring more machines here with the dbase problems. I lost over 500 workunits worth of credits the last time it crashed!! Way too often, for me, the inprogress and inconclusive numbers are the same or there are even more inconclusives than inprogress workunits and that scares me alot!! Too many other projects don't have those problems for MW to be STILL having them, someone needs to figure out how to ask for help.

mmonnin
Send message
Joined: 2 Oct 16
Posts: 102
Credit: 81,199,642
RAC: 18,711

Message 67053 - Posted: 10 Feb 2018, 13:07:58 UTC

It was down yesterday for a bit and I ran out of work while I couldn't access the servers. After running out of work my 280x ran 9 E@H tasks so it would down a couple of hours.

Profile ritterm
Avatar
Send message
Joined: 16 Jun 08
Posts: 92
Credit: 365,629,434
RAC: 0

Message 67076 - Posted: 13 Feb 2018, 16:38:16 UTC

This seems to be happening more and more recently, with multiple outages on some days.

Profile Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist
Send message
Joined: 25 Feb 13
Posts: 501
Credit: 34,647,251
RAC: 224

Message 67077 - Posted: 13 Feb 2018, 17:42:47 UTC

Hey Everyone,

I implemented something that might fix this today. The problem seems to be too many open connections on the database. I've made some configuration changed to help improve connection turnover and increase the connection limit.

Jake

EG
Send message
Joined: 13 Jul 13
Posts: 3
Credit: 714,495,223
RAC: 206,092

Message 67087 - Posted: 15 Feb 2018, 19:49:34 UTC - in response to Message 67077.

Hey Everyone,

I implemented something that might fix this today. The problem seems to be too many open connections on the database. I've made some configuration changed to help improve connection turnover and increase the connection limit.

Jake


Didn't solve my problem and now it's effecting Blackhawk 3

Hasn't effected either 1 or 4 yet (they are dual xeons while 2+3 are 8350's if that makes a difference)

I've set the boinc manager project priority to 100 and Collatz priority to 1 so at least they don't run idle anymore.

But seriously, this has gone to a very reliable project to a hit or miss proposition....

Don't know how much longer this can go....

Profile Keith Myers
Avatar
Send message
Joined: 24 Jan 11
Posts: 149
Credit: 102,237,241
RAC: 20,867

Message 67088 - Posted: 15 Feb 2018, 21:08:46 UTC

MilkyWay in backoff now because server database can't be opened message.
____________

EG
Send message
Joined: 13 Jul 13
Posts: 3
Credit: 714,495,223
RAC: 206,092

Message 67090 - Posted: 15 Feb 2018, 23:44:10 UTC - in response to Message 67077.

Hey Everyone,

I implemented something that might fix this today. The problem seems to be too many open connections on the database. I've made some configuration changed to help improve connection turnover and increase the connection limit.

Jake


I don't know what you did, but everything seems to be working fine now. Carrying a full load on everything with full cache's

And good traffic relay also no delay in d'loading or uploading....

Now if they validate we are Good to GO!

Profile Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist
Send message
Joined: 25 Feb 13
Posts: 501
Credit: 34,647,251
RAC: 224

Message 67091 - Posted: 16 Feb 2018, 14:54:19 UTC

Hey everyone,

I'm working on a hardware fix for the server. It's been running out of RAM, but I just ordered another 32gb of RAM for it. Hopefully that will get it running a little more smoothly. It should be here in a week or two.

Jake

Cautilus
Send message
Joined: 29 Jul 14
Posts: 9
Credit: 600,474,132
RAC: 1,186,101

Message 67092 - Posted: 16 Feb 2018, 15:54:04 UTC

Great to hear a permanent solution is on the way, thanks Jake!

Profile mikey
Avatar
Send message
Joined: 8 May 09
Posts: 2183
Credit: 232,361,889
RAC: 230,124

Message 67106 - Posted: 19 Feb 2018, 16:34:18 UTC - in response to Message 67091.

Hey everyone,

I'm working on a hardware fix for the server. It's been running out of RAM, but I just ordered another 32gb of RAM for it. Hopefully that will get it running a little more smoothly. It should be here in a week or two.

Jake


Is that the reason the d=base crashes like it did last night?
Validation inconclusive (479) for me now while it was around 150 or so. MOST of those are "unsent"!!

Profile Keith Myers
Avatar
Send message
Joined: 24 Jan 11
Posts: 149
Credit: 102,237,241
RAC: 20,867

Message 67124 - Posted: 22 Feb 2018, 2:38:10 UTC

Jake, I haven't seen a single 'database' error in any of my three crunchers today. Would have seen at least a couple before the RAM upgrade.

However,I still have 279 Inconclusive tasks listed and all of them have 'unsent' status for my wingmen.
____________

mmonnin
Send message
Joined: 2 Oct 16
Posts: 102
Credit: 81,199,642
RAC: 18,711

Message 67125 - Posted: 22 Feb 2018, 11:04:14 UTC

Doing much better. From nearly 4k to 1.4k. Most teams had a big step up on stats pages yesterday and as WUs were validated by the 2nd person.

Profile mikey
Avatar
Send message
Joined: 8 May 09
Posts: 2183
Credit: 232,361,889
RAC: 230,124

Message 67126 - Posted: 22 Feb 2018, 11:07:22 UTC - in response to Message 67125.

Doing much better. From nearly 4k to 1.4k. Most teams had a big step up on stats pages yesterday and as WUs were validated by the 2nd person.


Mine too!! I went from over 300 inconclusive with 60+ unsent to only 61 inconclusive this morning and zero unsent wu's!! WOO HOO!!

Max_Pirx
Send message
Joined: 13 Dec 17
Posts: 5
Credit: 85,292,650
RAC: 744,536

Message 67128 - Posted: 22 Feb 2018, 21:44:35 UTC

Yeah, it definitely looks better now, significant drop in my Unsent Wus. Nice one (^_^)

Profile Keith Myers
Avatar
Send message
Joined: 24 Jan 11
Posts: 149
Credit: 102,237,241
RAC: 20,867

Message 67129 - Posted: 23 Feb 2018, 0:30:38 UTC

RAM upgrade looks like a win for the project. I have no 'unsent' tasks in my Inconclusives.
____________

Profile Jake Weiss
Volunteer moderator
Project developer
Project tester
Project scientist
Send message
Joined: 25 Feb 13
Posts: 501
Credit: 34,647,251
RAC: 224

Message 67130 - Posted: 23 Feb 2018, 15:32:06 UTC

Awesome! Glad everything is looking better. We are actually doing another hardware upgrade late next week to put a second CPU into the server so hopefully it will be smooth sailing for a while after that.


Thanks for sticking with us everyone,

Jake


Post to thread

Message boards : Number crunching : MW@H DBase problems


Main page · Your account · Message boards


Copyright © 2018 AstroInformatics Group