you need to set the default zone as master zone. you can try: radosgw-admin zonegroup set < zg.json where the json is the json return from radosgw-admin zonegroup get with master_zone field set to "default" Orit On Mon, Jul 25, 2016 at 11:17 PM, Frank Enderle <frank.enderle@xxxxxxxxxx> wrote: > It most certainly looks very much like the same problem.. Is there a way to > patch the configuration by hand to get the cluster back in a working state? > > -- > > From: Shilpa Manjarabad Jagannath <smanjara@xxxxxxxxxx> > Date: 25 July 2016 at 10:34:42 > To: Frank Enderle <frank.enderle@xxxxxxxxxx> > Cc: ceph-users@xxxxxxxxxxxxxx <ceph-users@xxxxxxxxxxxxxx> > Subject: Re: Problem with RGW after update to Jewel > > > ----- Original Message ----- >> From: "Frank Enderle" <frank.enderle@xxxxxxxxxx> >> To: ceph-users@xxxxxxxxxxxxxx >> Sent: Monday, July 25, 2016 1:28:10 AM >> Subject: Problem with RGW after update to Jewel >> >> Hi, >> >> a while ago I updated a cluster from Infernalis to Jewel. After the update >> some problems occured, which I fixed (I had to create some additional pool >> which I was helped with in the IRC channel) - so the cluster now ran fine >> until we tried to add an addtional bucket. Now I get the following error >> in >> the error log: >> >> 2016-07-24 19:50:45.978005 7f6ce97fa700 1 ====== starting new request >> req=0x7f6ce97f4710 ===== >> 2016-07-24 19:50:46.021122 7f6ce97fa700 0 sending create_bucket request to >> master zonegroup >> 2016-07-24 19:50:46.021135 7f6ce97fa700 0 ERROR: endpoints not configured >> for >> upstream zone >> 2016-07-24 19:50:46.021148 7f6ce97fa700 0 WARNING: set_req_state_err >> err_no=5 >> resorting to 500 >> 2016-07-24 19:50:46.021249 7f6ce97fa700 1 ====== req done >> req=0x7f6ce97f4710 >> op status=-5 http_status=500 ====== >> 2016-07-24 19:50:46.021304 7f6ce97fa700 1 civetweb: 0x7f6dac001420: >> 10.42.20.5 - - [24/Jul/2016: 19:50:45 +0000 ] "PUT /abc/ HTTP/1.1" 500 0 - >> Cyberduck/4.7.3.18402 (Mac OS X/10.11.6) (x86_64) >> >> I already tried to fix the problem using the script at >> >> https://www.mail-archive.com/ceph-users@xxxxxxxxxxxxxx/msg28620.html >> >> with the outcome that all users disappeared and no bucket could be access. >> So >> I restored the backup .rgw.root and it now works again, but still I can't >> create buckets. Obviously something has been mixed up with the >> zone/zonegroup stuff during the update. >> >> Would be somebody able to take a look at this? I'm happy to provide all >> the >> required files; just name them. >> >> Thanks, >> >> Frank >> > > It looks like http://tracker.ceph.com/issues/16627, pending backport. > > >> _______________________________________________ >> ceph-users mailing list >> ceph-users@xxxxxxxxxxxxxx >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >> > > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |