I'm seeing some errors in S@H:

2004-09-28 02:35:47 [SETI@home] Unrecoverable error for result 27ap04aa.4270.16018.304840.5_0 (Maximum memory usage exceeded)

2004-09-28 07:45:48 [SETI@home] Unrecoverable error for result 27ap04aa.4270.17633.611064.95_0 (Maximum memory usage exceeded)

What's the deal with this?

[ Edited to add: The box in question has 1 GB RAM installed, 400 MB cached and only about 25 MB committed to the swap partition. I really, really doubt that the system actually used up 75% of the swap partition (which is the limit I've set for BOINC in my general prefs). ]

Posting to say 'me too'

I detached from seti to beta test LHC, and now coming back to Seti, I run into this error on my first work unit.

2004-10-02 23:48:30 [SETI@home] Throughput 55986 bytes/sec
2004-10-02 23:48:30 [SETI@home] Started download of setiathome_4.02_i686-pc-linu x-gnu
2004-10-02 23:48:30 [SETI@home] Finished download of background_1096755529
2004-10-02 23:48:30 [SETI@home] Throughput 70140 bytes/sec
2004-10-02 23:48:30 [SETI@home] Started download of 04mr04aa.22434.7729.273584.1 10
2004-10-02 23:48:37 [SETI@home] Finished download of 04mr04aa.22434.7729.273584. 110
2004-10-02 23:48:37 [SETI@home] Throughput 51543 bytes/sec
2004-10-02 23:49:13 [SETI@home] Finished download of setiathome_4.02_i686-pc-lin ux-gnu
2004-10-02 23:49:13 [SETI@home] Throughput 171796 bytes/sec
2004-10-03 00:38:04 [LHC@home] Computation for result v64lhc1000pro16s6_8567.41_1_sixvf_99650 finished
2004-10-03 00:38:04 [SETI@home] Starting result 04mr04aa.22434.7729.273584.110_0 using setiathome version 4.02
2004-10-03 00:38:04 [LHC@home] Started upload of v64lhc1000pro16s6_8567.41_1_sixvf_99650_2_0
2004-10-03 00:38:09 [LHC@home] Finished upload of v64lhc1000pro16s6_8567.41_1_sixvf_99650_2_0
2004-10-03 00:38:09 [LHC@home] Throughput 27385 bytes/sec
2004-10-03 01:38:04 [SETI@home] Pausing result 04mr04aa.22434.7729.273584.110_0 (removed from memory)
2004-10-03 01:38:04 [LHC@home] Starting result v64lhc1000pro16s6_8567.59_1_sixvf_99652_2 using sixtrack version 4.46
2004-10-03 01:38:09 [SETI@home] Aborting result 04mr04aa.22434.7729.273584.110_0: exceeded memory limit 33554422.000000
2004-10-03 01:38:09 [SETI@home] Unrecoverable error for result 04mr04aa.22434.7729.273584.110_0 (Maximum memory usage exceeded)
2004-10-03 01:38:09 [SETI@home] Unrecoverable error for result 04mr04aa.22434.7729.273584.110_0 (Maximum memory usage exceeded)
2004-10-03 01:38:09 [SETI@home] Deferring communication with project for 1 minutes and 0 seconds
2004-10-03 01:38:09 [SETI@home] Deferring communication with project for 1 minutes and 0 seconds
2004-10-03 01:38:10 [SETI@home] Computation for result 04mr04aa.22434.7729.273584.110 finished
2004-10-03 01:38:13 [SETI@home] Started upload of 04mr04aa.22434.7729.273584.110_0_0
2004-10-03 01:38:15 [SETI@home] Finished upload of 04mr04aa.22434.7729.273584.110_0_0
2004-10-03 01:38:15 [SETI@home] Throughput 64803 bytes/sec
2004-10-03 01:39:10 [---] Insufficient work; requesting more
2004-10-03 01:39:10 [SETI@home] Requesting 16521 seconds of work

I may have some further light to shed on this issue. This afternoon, I noticed that system load was up at 2.00 instead of the usual 1.00. When I took a look at top, both LHC and SETI were crunching away. Later, system load was back at 1.00 and my BOINC error log shows:

2004-10-26 17:53:59 [SETI@home] Unrecoverable error for result 05mr04aa.22210.25440.809642.134_1 (Maximum memory usage exceeded)

I'm guessing that the memory usage/WU failure and the simultaneous project execution are related.

---

I had the same problem. It went away (I hope) after I set
"Leave applications in memory while preempted?" to "no"

Regards Hans

P.S:
The boinc client seems to have problems when running more than one project,
or more then one client on a HT CPU

---

> I had the same problem. It went away (I hope) after I set
> "Leave applications in memory while preempted?" to "no"

Interesting. The problem for me is that setting causes me to throw away ~20 min. of CPU time for CDPN due to its infrequent checkpoints.

Thanks for the post; it definitely seems as though the problem is related to simultaneous project execution on the same CPU/core.

---