Re: [pve-devel] [Patch V2 guest-common] fix #1694: Replication risks permanently losing sync in high loads due to timeout bug

> ;)) > > > +}; > > > > -die $err if $err; > > +# old snapshots will removed by next run from prepare_local_job. > > +warn $@ if $@; > > +$logfunc->("delete stale replication snapshot error: $@") if $@; > > again, is this double warn + log needed? (I ask because 2 out of 3

Re: [pve-devel] [Patch V2 guest-common] fix #1694: Replication risks permanently losing sync in high loads due to timeout bug

minor nit: please use a short subject that indicates what the commit does, not what the bug was that it fixes. e.g., something like fix #1694: make failure of snapshot removal non-fatal On Thu, Apr 12, 2018 at 09:46:03AM +0200, Wolfgang Link wrote: > If the pool is under heavy load ZFS will low

Re: [pve-devel] [Patch V2 guest-common] fix #1694: Replication risks permanently losing sync in high loads due to timeout bug

> Dietmar Maurer hat am 12. April 2018 um 11:06 > geschrieben: > > > > diff --git a/PVE/Replication.pm b/PVE/Replication.pm > > index 9bc4e61..d8ccfaf 100644 > > --- a/PVE/Replication.pm > > +++ b/PVE/Replication.pm > > @@ -136,8 +136,18 @@ sub prepare { > >

Re: [pve-devel] [Patch V2 guest-common] fix #1694: Replication risks permanently losing sync in high loads due to timeout bug

> @@ -293,12 +303,16 @@ sub replicate { > die $err; > } > > -# remove old snapshots because they are no longer needed > -$cleanup_local_snapshots->($last_snapshots, $last_sync_snapname); > +eval { > + # remove old snapshots because they are no longer needed > +

Re: [pve-devel] [Patch V2 guest-common] fix #1694: Replication risks permanently losing sync in high loads due to timeout bug

> diff --git a/PVE/Replication.pm b/PVE/Replication.pm > index 9bc4e61..d8ccfaf 100644 > --- a/PVE/Replication.pm > +++ b/PVE/Replication.pm > @@ -136,8 +136,18 @@ sub prepare { > $last_snapshots->{$volid}->{$snap} = 1; > } elsif ($snap =~ m/^\Q$prefix\E/) { >