You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm having issues restarting influxDB after it had run for almost 2 weeks. I'm running influxDB v0.9.6.1 on a Raspberry Pi 2 (4 ARM cores, 1 GB internal memory). The database files are around 733 MB in total. I'm using telegraf to collect stats at a 30s frequency.
I get the following error immediately (from the influxdb.log):
2016/01/12 11:54:41 InfluxDB starting, version 0.9, branch unknown, commit unknown, built unknown
2016/01/12 11:54:41 Go version go1.5.2, GOMAXPROCS set to 4
2016/01/12 11:54:41 Using configuration at: /etc/influxdb/influxdb.conf
[metastore] 2016/01/12 11:54:41 Using data dir: /pi/influxdb/meta
[metastore] 2016/01/12 11:54:41 Node at localhost:8088 [Follower]
[metastore] 2016/01/12 11:54:41 Skipping cluster join: already member of cluster: nodeId=1 raftEnabled=true peers=[localhost:8088]
[metastore] 2016/01/12 11:54:43 Node at localhost:8088 [Leader]. peers=[localhost:8088]
[metastore] 2016/01/12 11:54:43 spun up monitoring for 1
[metastore] 2016/01/12 11:54:43 Updated node id=1 hostname=localhost:8088
[store] 2016/01/12 11:54:43 Using data dir: /pi/influxdb/data
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/14
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/15
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/16
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/17
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/19
[wal] 2016/01/12 11:54:43 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:43 WAL writing to /pi/influxdb/wal/_internal/monitor/21
[wal] 2016/01/12 11:54:44 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:44 WAL writing to /pi/influxdb/wal/_internal/monitor/22
[wal] 2016/01/12 11:54:44 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:44 WAL writing to /pi/influxdb/wal/_internal/
monitor/23
[wal] 2016/01/12 11:54:44 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:44 WAL writing to /pi/influxdb/wal/nodered/default/20
[wal] 2016/01/12 11:54:44 WAL starting with 30720 ready series size, 0.50 compaction threshold, and 52428800 partition size threshold
[wal] 2016/01/12 11:54:44 WAL writing to /pi/influxdb/wal/telegraf/default/10
[retention] 2016/01/12 11:54:44 retention policy enforcement terminating
[monitor] 2016/01/12 11:54:44 shutting down monitor system
[handoff] 2016/01/12 11:54:44 shutting down hh service
[metastore] 2016/01/12 11:54:44 RPC listener accept error and closed: network connection closed
[metastore] 2016/01/12 11:54:44 exec listener accept error and closed: network connection closed
[subscriber] 2016/01/12 11:54:44 closed service
run: open server: open tsdb store: failed to open shard 18: new engine: cannot allocate memory
Going through the logs i find this error, before systemd tries to restart influxdb:
I tried 0.10.0-beta1 with go 1.4.3, but that doesn't seem to solve the issue. I removed the database in question and restarted. Lets see if it fails again after two weeks ;-)
1GB of memory is likely not enough for the your dataset. Increasing your RAM should avoid the cannot allocate memory.
The fatal error: fault panic is related to the bz1 engine which is deprecated and replaced by the tsm engine in 0.10. Issues with the bz1 engine will not be fixed as the solution is to upgrade to 0.10 and migrate shards to tsm using the conversion tool.
Hi all,
I'm having issues restarting influxDB after it had run for almost 2 weeks. I'm running influxDB v0.9.6.1 on a Raspberry Pi 2 (4 ARM cores, 1 GB internal memory). The database files are around 733 MB in total. I'm using telegraf to collect stats at a 30s frequency.
I get the following error immediately (from the influxdb.log):
Going through the logs i find this error, before systemd tries to restart influxdb:
Any idea what to do to fix this?
The text was updated successfully, but these errors were encountered: