This is no longer happening. I started to prep for the downtime this morning and in the process realized I didn't actually need such a large downtime. I brought down the server for 2 minutes at 8 am this morning and got the data out.
On Mon, May 23, 2016 at 11:37 AM, Nigel Babu <[email protected]> wrote: > I was trying to clone gluster this morning and noticed that it was awfully > slow and then completely stopped responding. I've restarted gerrit and it > seems to work much better. > > This has proven that my monitoring of gerrit works though, so we can, in > the future, know of these outages before someone reports them. > > For future reference > 1) A HTTP call to the gerrit API will fail if gerrit is unhappy. > 2) We're limited by CPU rather than RAM when gerrit doesn't respond. This > may help us debug what we're doing wrong with our setup. > > -- > nigelb > -- nigelb
_______________________________________________ Gluster-infra mailing list [email protected] http://www.gluster.org/mailman/listinfo/gluster-infra
