Would it be possible to test batches in lasieved before submitting to the grid? There have been two recent batches that all contain computation errors and cause my machines to become unreliable and then only get 1 WU per day per quota. Just as they become reliable again, another batch of ~2000 WUs get submitted and all contain errors again and I'm back to unreliable again. To make the situation even worse, nobody recognizes that the batch is totally messed up and when the errors are returned to the server another erroneous WU is submitted again. This work isn't labeled as beta!

Respectfully request:

1. Test before submitting batch
2. Monitor the batch after submission and if erroneous, cancel remaining work

Edited to add error data:
<core_client_version>8.0.4</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)</message>
<stderr_txt>
boinc initialized
work files resolved, now working
-> ../../projects/escatter11.fullerton.edu_nfs/lasieved_1.11_x86_64-pc-linux-gnu
-> -r
-> -f
-> 42432000
-> -c
-> 16000
-> -R
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1.poly
-> -o
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1_42432_0_r1249026950_0
Expected to create 12 schedules on side 0, have 7
09:43:32 (35873): called boinc_finish(1)

</stderr_txt>
]]>

Linux bad.

<core_client_version>7.22.2</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)</message>
<stderr_txt>
boinc initialized
work files resolved, now working
-> ../../projects/escatter11.fullerton.edu_nfs/lasieved_1.11_x86_64-pc-linux-gnu
-> -r
-> -f
-> 48624000
-> -c
-> 16000
-> -R
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1.poly
-> -o
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1_48624_1_r315520415_0
Expected to create 12  schedules on side 0, have 7
05:49:51 (2634649): called boinc_finish(1)

</stderr_txt>
]]>

Windows good.

<core_client_version>8.0.2</core_client_version>
<![CDATA[
<stderr_txt>
boinc initialized
work files resolved, now working
-> projects/escatter11.fullerton.edu_nfs/lasieved_1.08_windows_intelx86.exe
-> -r
-> -f
-> 48624000
-> -c
-> 16000
-> -R
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1.poly
-> -o
-> ../../projects/escatter11.fullerton.edu_nfs/177473_41m1_48624_3_r440598530_0

total yield: 24126, q=48640001 (0.22471 sec/rel,  100.00625 % done of 16000)called boinc_finish

</stderr_txt>
]]>

Why Linux is bad and Windows is fine? Workunit https://escatter11.fullerton.edu/nfs/workunit.php?wuid=355476863.