Joined: 15 Jul 99
I have no idea why, but at 2:13 AM last night, so 47 min (well, technically 1h47 min) shy of 8 days after it was correctly done the first time, my computer decided to put the time forward by yet another hour (and actually changed the hour, according to logs, not just the offset). Wasn't using it just then, got to it soon after, didn't realize it, but at 3:03 AM (so 50 min later but another weird time otherwise) it synced with the time server and fixed itself. Normally the sync is at 1 AM, and once a week, but now hadn't had a successful one since March 12 for some reason.
Anyway, when it fixed itself my connection reset twice, and apparently some of that confused BOINC. Didn't notice it right away, but some 45-50 min later I checked running processes and saw CPU not at 100%, just one WU running (it's a dual core, so should be 2). Check BOINC manager, shows 2 running, but not making any progress. Suspend them, they show up suspended, but not logged as so, no others start, and in running processes nothing changes. Try to close BOINC Manager (BOINC is installed as service), say to stop running WUs too, I think the manager itself closes but the other processes, including the service controller one just started and the WU, stay open. Also see "A timeout (30000 milliseconds) was reached while waiting for a transaction response from the BOINC service." in system logs.
So force them all closed, restart BOINC manager, boinctray doesn't start (but it shows up in tray normally, so no idea what that does), start it as well manually just in case... See one S@h WU error out immediately. Whether it was the one I forced closed or one which should have been finished during that time it was messed up, before I noticed it, don't know. In logs I see one connection just when the time skipped forward and one more some minutes later, nothing reported or downloaded on either, nothing else till I restarted it, not even an exit message when I tried to turn it off. Seems to be working normally since (also rebooted after that), but benchmarks report an integer speed almost 2% lower than it should be. And steady, ran benchmarks several times with results within a few points of each other. (Nope, no updates installed or Meltdown/Spectre mitigations enabled then, nothing done on that front since Mar 14.)
So what was that all about?
©2019 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.