Second replication begins if first replication is not finished

[woodsb02]

During first replication of many Gigabytes of data, I initially had the interval of the pull job set as 10m, and the first replication would not be finished by the time the second one was called to start. I checked the status many hours later and could see numerous ssh sessions running which led me to believe multiple replication jobs were now running at once (which I dont think should ever happen). I expected that if another replication job was called to start before the previous had finished, the new job would just be cancelled entirely.

I did not look into the state of my replicated data, or if the replications were proceeding ok. It was purely the fact that multiple zrepl ssh sessions were running that led me to believe this was the behaviour.

[problame]

This is a bug. WIll fix.

[problame]

So I guess this was a pull job? I cannot reproduce the issue.

cmd/config_job_pull.go:94 JobStart() is strictly sequential and will not reconnect unless the previous pull finished.

What's still an open issue: where did all the dangling ssh sessions come from?

[woodsb02]

I will try to replicate it this weekend, and will report back on my findings. I didn’t spend the time to investigate and record my findings last time... I just remember seeing numerous lines from “sudo pgrep -lf zrepl”

[problame]

Were you able to replicate the described behavior?

[problame]

OK, I was able to observer the issue on a testing system. I saw lots of defunct processes, most likely ssh processes that timed out but were not waitpid() for by zrepl.
Sadly, the logs are gone because the testing system was also used to test TCP logger, which doesn't handle timeouts on the connection well, see #26

[problame]

So I think I fixed the issue in 6b5bd0a --- it just landed in zrepl master.
Are you in a situation where you can just build zrepl master and check if the issue is resolved?

[woodsb02]

Hi Christian,
Yes, you are correct - this issue was the same as the one reported in #56.
I have just finished testing with the new latest (unreleased) version of zrepl, and can confirm this is now fixed.
Thanks for your work on fixing this!
Cheers,
Ben