|| at Feb 11, 2015 at 7:57 pm
Well, the build coordinator does auto-restart on crash and the VM
auto-restarts on crash and is auto-migrated around by GCE if the machine
It's a single point of failure, but it's a highly available single point of
failure, as opposed to a machine that gets unplugged. It's true it had the
fd leak before and got wedged so bugs can happen, but we've had no problems
with its availability besides that, which was fixed.
I asked Shawn yesterday when Gerrit will support pushing to Github natively
but apparently it's stalled behind other stuff again.
On Wed, Feb 11, 2015 at 11:47 AM, Russ Cox wrote:
Sorry, the machine that was doing the syncing got disconnected from the
network over the weekend and needed to be unplugged and replugged in (I
don't know the details). It's all better now. We may move this into the
builder coordinator but it's still a single point of failure.
On Mon, Feb 9, 2015 at 7:38 PM, Brad Fitzpatrick wrote:
Google's gerrit instances run in a different environment. That plugin
doesn't work as-is.
On Feb 9, 2015 4:34 PM, "minux" wrote:
Gerrit has an official plugin called replication that does this
We don't need to run another job to sync the repos (and even more
the replication plugin supports force updating on the github mirror, so
someone accidentally pushes commit to the github repo, the next gerrit
will wipe out that commit)https://gerrit-review.googlesource.com/#/admin/projects/plugins/replication
I use this plugin my Gerrit instance to push updates to github, and it
You received this message because you are subscribed to the Google Groups "golang-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to email@example.com.
For more options, visit https://groups.google.com/d/optout.