A weird sleep timing issue... or a problem with latest ATI version?
Yesterday I updated ATI 2017 to Build 8058. It seemed to go well.
I discovered this morning that my 7:00 PM daily backup did not go well last night. Looking at the schedul2 logs, I found this:
23.06.2017 18:58:01.815 0000000000000EE8 OnServiceMainWake: next time = 23.06.2017 19:00:00
23.06.2017 18:58:01.815 0000000000002354 Host poller wait for 4294967295 ms
23.06.2017 18:58:01.850 00000000000008F0 OnServiceEvent(SE_LOGON_REP)
23.06.2017 18:58:01.850 00000000000008F0 GetCurrentSession: WtsId=1 LogonSid=S-1-5-5-0-143204 User=ZOMO\BRUNO
23.06.2017 18:58:01.851 00000000000008F0 Session 1 state is 0 (WTSActive)
23.06.2017 18:58:01.851 00000000000008F0 Session 1 flags is 1 - unlocked
23.06.2017 18:58:01.851 00000000000008F0 Created interactive session
23.06.2017 18:58:01.851 0000000000001C48 Runner wait for 4294967295 ms
23.06.2017 18:59:54.343 00000000000008F0 OnServiceEvent(SE_POWER)
23.06.2017 18:59:54.343 00000000000008F0 power event 0x4 (PBT_APMSUSPEND) received
23.06.2017 18:59:54.372 00000000000000A0 PowerRegisterSuspendResumeNotification callback: Type = 0x4 (PBT_APMSUSPEND)
23.06.2017 19:41:53.637 000000000000146C OnServiceTimeChange: suspend
23.06.2017 19:41:53.637 000000000000146C Time zone: UTC-7.00 (Pacific Daylight Time)
23.06.2017 19:41:53.726 00000000000000A0 PowerRegisterSuspendResumeNotification callback: Type = 0x7 (PBT_APMRESUMESUSPEND)
23.06.2017 19:41:53.726 0000000000000EE8 OnServiceMainWake: begin
23.06.2017 19:41:53.727 0000000000000EE8 Trying task 1-22 as TIME_NORMAL
I always restart the scheduler 2 minutes before the backup. We can see from the first line here that it should wake up at 19:00 to start the backup. But then it appears that 6 seconds before that time, the system went to sleep and did not wake up until 19:41, at which time it started the backup. And yes, I do have the option checked to wake up a sleeping computer. It just so happened that it was time to do a full backup and my NAS automatically shuts down at 20:00. I think the full backup was within a minute of finishing when the NAS shut down. As a result I could see ATI retrying this morning. I turned on the NAS and the backup went fine.
Is the problem something new with the new build? Is there a timing window where if the computer is going to sleep at the time the backup is supposed to start then it will fail? I'll be watching this closely.


- Log in to post comments

I've disabled the restart task so we'll see how things go.
- Log in to post comments

Did not go well. My 7:00 pm backup did not run, although the Schedul2 log shows the scheduler trying to start that pesky ga_service /ransomware task. I forgot to disable the ga_service stuff after downloading the latest build. I don't even know why it wants to run it...I don't have the NG version.
I'm going back to my previous plan.
- Log in to post comments

Bruno, do you still have an open support case for the schedule issue, if so, I would feed this latest issue into that case, including why you are seeing the ransomware task getting involved when you have the Standard version?
- Log in to post comments

The support case is no longer open. Apparently they don't like to keep cases open unless they are actively researching them. Because I had a workaround, they wanted to close it.
Whenever there is an update, there are two ga_service tasks that get scheduled. The one with the /ransomware argument never succeeds so it just keep retrying. Using the Acronis Scheduler Manager, I'm able to disable them easily.
I wish I had the source code. I spent 30 years as a systems software developer; I could probably debug it from the code faster than all the back and forth I've been doing.
- Log in to post comments

Bruno, even though your support case is 'closed' I would send off an update about this problem to them to say that your workaround is not working 100% and you would still like the issue to be investigated and resolved!
- Log in to post comments

Steve, when they closed the issue on the sleep/wake problem, they informed me that the fix would be in the ATI 2018 version so there was no expectation that it would be fixed in this build.
I'm not sure if this is related, or something different. I will pass on the Schedul2 log to them anyway.
- Log in to post comments

Bruno, it may be worth asking Slava or Ekaterina if they would consider you for early adopter testing of the ATI 2018 version when they decide to start this for users?
- Log in to post comments