On all my Odroid-N2 the Eprot_v1 end after 1d3h30m runtime with 197 (0x000000C5) EXIT_TIME_LIMIT_EXCEEDED.

<core_client_version>7.16.6</core_client_version>
<![CDATA[
<message>
exceeded elapsed time limit 99186.72 (1000000.00G/10.08G)</message>
<stderr_txt>
04:31:39 (1302552): wrapper (7.17.26016): starting
04:31:39 (1302552): wrapper (7.17.26016): starting
04:31:39 (1302552): wrapper: running cmdock (-c -r target.prm -p "/var/lib/boinc-client/slots/3/data/scripts/dock.prm" -f htvs.ptc -i ligands.sdf -o docking_out)

</stderr_txt>
]]>

Have you changed the settings? All host run this before without problem.
https://www.sidock.si/sidock/results.php?hostid=12224

Hello JagDoc!

Thank you for problem reporting! Current bound set now at 1E+15 GFLOPS. If CPU Performance estimate as 2 GFLOPS/s, time limit must be a 500 000 second. But BOINC client, as we seen, use a 10.08 GFLOPS/s estimate mark (may be additional coefficients influent to this). I increased bound up to 1E+16, but it will work only for new workunits.

Thank you.
I aborted on the ARMs the Eprot_v1.
Tomorrow we will se how it works with the new WUs

New WUs also ended with this error.

My 2400GE has the same problem with Eprot_v1 wu.
https://www.sidock.si/sidock/result.php?resultid=20168126

From where takes the app this value?
If you look in the resultid there is max. FLOPS des Gerätes 5.10 GFLOPS
and in the stderr: exceeded elapsed time limit 39481.06 (1000000.00G/25.30G)

Some of my team mates has the same Problem.
Please stop sending out Eprot_v1 untill problem is solved.

That's an old work unit - it was built on September 16 while the change only took place on September 17.
As hoarfrost stated, the change was only for new work units. Any work unit built before the change took place has a risk of having the same error.

This must be wrong settings in the wu.
Boinc shows on Eprot_v1 and 3CLpro wu the same estimated runtime.

They run on the same app so BOINC can't differentiate between the two. One could take 50 seconds and the other could take 24 hours and BOINC would basically average that out and present as an estimate because it's the same application.


Once the supply of pre -17 September time limits change work units is exhausted, the only ones left will be the ones with the larger time limits that shouldn't error out on your slower devices.