-
Type: Bug
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: 3.6.0-rc0
-
Component/s: Replication, Storage
-
Fully Compatible
-
ALL
-
v3.6
-
Storage 2018-02-12, Storage NYC 2018-03-12
-
(copied to CRM)
Simple insert loop running against 1-node replica set
db.c.insert({_id: 0, i: 0})
for (i = 0; i < 100000; i++)
db.c.update({_id: 0}, {$inc: {i: 1}})
Shows a large increase in journal syncs, disk write requests, bytes written. 3.4.9 left, 3.6.0-rc0 right:
It appears we're doing a flush on every update, even though there's no write concern. In the simple repro above this doesn't seem to affect performance (for reasons tbd) but with more complex workloads the additional journal i/o will impact performance.
Does not reproduce with standalone mongod.
- is duplicated by
-
SERVER-33537 High write IO after 3.6 upgrade
- Closed
- is related to
-
SERVER-33812 First initial sync oplog read batch fetched may be empty; do not treat as an error.
- Closed
-
WT-3531 Provide a way to flush the log if commits were out of timestamp order
- Closed
- related to
-
SERVER-36221 [3.6] Performance regression on small updates to large documents
- Closed
-
SERVER-37233 Increase in disk i/o for writes to replica set
- Closed