How can we increase n1ql timeout in the sync gateway?

Mobile Sync Gateway
1

We are running into the 75s timeout when the sync gateway is querying the server for changes. We have updated the databases.foo_db.view_query_timeout_secs to 300 since we are attempting load a large amount of documents, but the timeout is still happening at 75s. Are there any other updates that are needed?

These errors are logged to the sync gateway error logs:
[WRN] Error when querying index using statement: [SELECT meta({db}).xattrs._sync.sequence AS seq, meta({db}).xattrs._sync.rev AS rev, meta({db}).xattrs._sync.flags AS flags, META({db}).id AS id FROM {db}WHERE meta({db}).xattrs._sync.sequence >= $startSeq AND meta({db}).xattrs._sync.sequence < $endSeq AND META().id NOT LIKE ‘\_sync:%’] – base.(CouchbaseBucketGoCB).Query() at bucket_n1ql.go:63
[WRN] MultiChangesFeed got error reading changes feed "": [1080] Timeout 1m15s exceeded – db.(*Database).SimpleMultiChangesFeed.func1() at changes.go:493

Thanks for the help!

2

Hi @jblackburn,

There’s a bug in current versions of Sync Gateway that makes the view_query_timeout configuration option ineffective, and capped at 75 seconds (see issue: #3839)

This has been fixed in an upcoming SG 2.5.0 version, which we hope to be releasing soon.

3

Thanks for the update. We’ll be watching for the new release.