I just discoverd these strange "update_mapping" loglines come from a completely unrelated thing, so please take this post as invalid and accept my apologies.
On Thursday, June 19, 2014 1:21:32 PM UTC-4, JoeZ99 wrote: > > This is a somehow bizarre question. I really hope somebody jumps in, > because I'm losing my mind. > > We've set a system by which our one-machine cluster gets updated indexes > that have been made in other clusters by restoring snapshots. > > Long story short: > > for a few hours, the cluster is restoring snapshots, each one of them > containing information about two indexes. of course , the "global_state" > flag is set to false, because we don't want to recover the cluster, just > those two indexes. > > Say during those few hours , the cluster have restored about 500 > snapshots, one after another (there is never two restore processes at the > same time). By looking at the logs, it goes flawlessly : > > > > [2014-06-19 00:00:01,318][INFO ][snapshots ] [Svarog] > restore [backups-1:5e51361312cb68f41e1cb1fa5597672a_ts20140618235915350570 > ] is done > [2014-06-19 00:00:02,363][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:08,653][INFO ][cluster.metadata ] [Svarog] [ > 5e51361312cb68f41e1cb1fa5597672a_ts20140617220817522348] deleting index > [2014-06-19 00:00:09,286][INFO ][cluster.metadata ] [Svarog] [ > 5e51361312cb68f41e1cb1fa5597672a_phonetic_ts20140617220817904810] > deleting index > [2014-06-19 00:00:09,815][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:15,570][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:15,938][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:16,208][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:20,669][INFO ][snapshots ] [Svarog] > restore [backups-1:70e3583358803e70dc60a83953aaca9e_ts20140618235930121779 > ] is done > [2014-06-19 00:00:21,585][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 00:00:26,992][INFO ][cluster.metadata ] [Svarog] [ > 70e3583358803e70dc60a83953aaca9e_ts20140617220848057264] deleting index > [2014-06-19 00:00:27,601][INFO ][cluster.metadata ] [Svarog] [ > 70e3583358803e70dc60a83953aaca9e_phonetic_ts20140617220848563815] > deleting index > > after restoring the snapshot, outdated version of the indices are removed > (because the indices recovered from the snapshot are newer). > > this goes quite well, and there is no significant load on the machine > while doing this. > > but, at some poing, the cluster starts to issue "udpate_mapping" commands > with no apparent reason (I'm almost sure there's been no interaction from > outside)... > [2014-06-19 04:38:36,293][INFO ][snapshots ] [Svarog] > restore [backups-1:99cbf66451446e6fe770878e84b4349b_ts20140619043819745139 > ] is done > [2014-06-19 04:38:37,238][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 04:38:44,016][INFO ][cluster.metadata ] [Svarog] [ > 99cbf66451446e6fe770878e84b4349b_ts20140604042653951289] deleting index > [2014-06-19 04:38:44,517][INFO ][cluster.metadata ] [Svarog] [ > 99cbf66451446e6fe770878e84b4349b_phonetic_ts20140604042655159506] > deleting index > [2014-06-19 05:57:24,721][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 05:57:34,869][INFO ][repositories ] [Svarog] > update repository [backups-1] > [2014-06-19 05:57:35,234<span style="color: #660;" class="styled > ... -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/949304b9-eba4-4328-badf-00f8288c36a3%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
