[13:19:02] <iksik> i have a quick question, is it true that oplog window size is somehow affecting secondary node full initial sync?
[13:33:23] <GothAlice> iksik: Yes and no. If a complete initial replication takes longer than the oplog window size, then the data transferred can not be "caught up" with the oplog, so initial replication fails.
[13:34:30] <GothAlice> (Transfer all data… can't catch up… transfer all data…)
[13:39:56] <iksik> yea, so that might be a reason for it to fail in here ;/ it correlates, fail after ~3days vs 3day oplog window, crapz
[13:40:45] <iksik> so what exactly happens when it goes outside of that window? because my mongo node gets killed with OOM
[13:41:39] <iksik> after 3 days virtual memory usage sky rockets (according to opsmanager)
[14:04:33] <GothAlice> That sounds unusual. If it detects there wasn't enough window (after completing the initial bulk transfer attempt), it should exit with an error describing the problem. Not… OOM.
[15:16:04] <iksik> GothAlice: thanks for confirmation
[23:00:24] <blottoface> I'm trying to export a collection with about five million records in it. Each record has about 14 fields. I want to export only the _id field from the collection into a single file
[23:01:16] <blottoface> Currently I'm using mongoexport and it's taking me about 9 minutes to export the entire collection. From there, I have to pare it down to only the _id field.
[23:02:01] <blottoface> A) does anyone here have an idea of how long it should take to export a collection like that?
[23:02:19] <blottoface> B) How can I pare it down to just the _id field?
[23:02:26] <blottoface> C) How could I do this faster?