I've gotten this message 4 times recently.
Here's the most recent one.
----------
7/11/2010 00:45:20 malariacontrol.net Aborting task wu_740_526_222417_0_1278706208_1: exceeded disk limit: 411.74MB > 381.47MB
-----------
I had been allocating 1.3 gigs to Boinc projects. After getting a couple of these messages I upped the allocation to a generous 10 gigs but after doing that I got the last message.
According to BOINC Malaria is using about 400 meg and I have over 9.5 GIGs free for BOINC projects. So what's up?
Currently I'm processing 4 Malaria WUs at a time and a SETI WU on a GPU.
/ 
- Reply
- Quote
I've gotten this message 4 times recently.
Here's the most recent one.
----------
7/11/2010 00:45:20 malariacontrol.net Aborting task wu_740_526_222417_0_1278706208_1: exceeded disk limit: 411.74MB > 381.47MB
-----------
I had been allocating 1.3 gigs to Boinc projects. After getting a couple of these messages I upped the allocation to a generous 10 gigs but after doing that I got the last message.
According to BOINC Malaria is using about 400 meg and I have over 9.5 GIGs free for BOINC projects. So what's up?
Currently I'm processing 4 Malaria WUs at a time and a SETI WU on a GPU.
It is a workunit setting not something we can change, they are having problems and I thought this was fixed a week or so ago. Are you crunching the version 6.42 workunits?
/
- Reply
- Quote
It is a workunit setting not something we can change, they are having problems and I thought this was fixed a week or so ago. Are you crunching the version 6.42 workunits?
Yup. This is the most recent WU. Looks like 6.42.
7/11/2010 10:27:19 malariacontrol.net Starting task wu_740_515_222418_0_1278706391_0 using openMalariaA version 642
It seems to be happening more and more often. I'm getting it on 15-20% of the WUs in the last day.
/
- Reply
- Quote
It is a workunit setting not something we can change, they are having problems and I thought this was fixed a week or so ago. Are you crunching the version 6.42 workunits?
Yup. This is the most recent WU. Looks like 6.42.
7/11/2010 10:27:19 malariacontrol.net Starting task wu_740_515_222418_0_1278706391_0 using openMalariaA version 642
It seems to be happening more and more often. I'm getting it on 15-20% of the WUs in the last day.
Two things, 1st you are using an older version of Boinc, the newest is 6.10.56, it is better and may handle the errors better. 2nd you have a LARGE cache for a project that is only giving out test units, MC is still testing this year and no units have been given out yet to anyone not signed up for the testing. ALL units are STILL being used to help stop the spread of Malaria but since they are testing, the units could all crash. You have units given to you back on July 9th, today is the 12th, that is a long time in testing. I checked all my pc's and since July 9th I have not had a unit crash on me, so once you get thru these you should be okay. On July 9th I had 5 units crash for the same reasons you are seeing, lack of memory. I have a 0.25 day cache with an additional 0.75 days extra. Meaning my cache is only about 1 days worth of units and I have 5 pc's currently crunching for MC. If you look at my pc's you can see that all that are doing MC is up to the latest version of Boinc.
/
- Reply
- Quote
A "Disk Limit" is set for each workunit we send. So, although you have allocated more disk space for the boinc projects, the workunits will be aborted when the authorized Disk Limit for these workunits is exceeded.
The Disk Limit is set as low as possible to allow a large part of hosts to compute our workunits.
We have now reduced the size of some types of workunits (we know that some are more disk-hungry than others) and increased the disk limit. This should solve these issues.
Thank you for your feedback.
____________
Guillaume Gnaegi
Swiss Tropical and Public Health Institute
http://www.swisstph.ch
/
- Reply
- Quote
A "Disk Limit" is set for each workunit we send. So, although you have allocated more disk space for the boinc projects, the workunits will be aborted when the authorized Disk Limit for these workunits is exceeded.
The Disk Limit is set as low as possible to allow a large part of hosts to compute our workunits.
We have now reduced the size of some types of workunits (we know that some are more disk-hungry than others) and increased the disk limit. This should solve these issues.
Thank you for your feedback.
ALRIGHT! Thanks!!
/
- Reply
- Quote
I've got one of these today:
https://malariacontrol.net/result.php?resultid=61459234
Task was only issued yesterday, so there are still problems with this issue.
____________
The Scottish BOINC Team Forum
/
- Reply
- Quote