>Jul 28 21:20:40 ip-172-31-2-161 kernel: [242909.768230] 
blk_update_request: I/O error, dev nvme0n1, sector 435160504

This looks like a hardware error. Can you move the discussion over 
to https://community.influxdata.com/ ?  Bonus points if you can provide a 
script to reproduce the problem.

On Friday, July 28, 2017 at 5:40:56 PM UTC-7, [email protected] wrote:
>
> Hi there 
>
> I am running InfluxDB on a single instance 8 core VM with 60GB RAM. The VM 
> is running Ubuntu 16.04 Xenial and InfluxDB 1.3.1.
>
> I am generating a write workload of about 2800/s. Each point in the 
> timeseries is a separate request and has between 50 and 300 fields. There 
> are about 10 measurements in the database.  
>
> A couple of hours after the workload started, the db crashed and I see the 
> following error in the logs. Once the compaction started, CPU usage shot up 
> and after a little while, the database shutdown with a sigpanic. I tried 
> restarting the database but the tsm files seem to have gotten corrupted. I 
> could only bring the database back up after deleting the tsm files.
>
> I am using the default influxdb configuration and I am currently not 
> batching the requests and will try to make a change to see if that helps. 
> Aside from this, I was wondering if there is any setting I could tweak to 
> prevent this from happening again or is the non-batch writes that I am 
> doing the main problem ?
>
> Any insight into this would be greatly appreciated
>
>  
> Jul 28 20:40:06 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:06Z 
> beginning full compaction of group 0, 2 TSM files engine=tsm1
> Jul 28 20:40:06 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:06Z 
> compacting full group (0) 
> /mnt/influxdb/data/blueshift/autogen/2/000000615-000000005.tsm (#0) 
> engine=tsm1
> Jul 28 20:40:06 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:06Z 
> compacting full group (0) 
> /mnt/influxdb/data/blueshift/autogen/2/000000637-000000004.tsm (#1) 
> engine=tsm1
> Jul 28 20:40:07 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:07Z 
> compacted level 1 group (0) into 
> /mnt/influxdb/data/blueshift/autogen/2/000000646-000000002.tsm.tmp (#0) 
> engine=tsm1
> Jul 28 20:40:07 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:07Z 
> compacted level 1 3 files into 1 files in 12.705450299s engine=tsm1
> Jul 28 20:40:29 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:29Z 
> compacted full group (0) into 
> /mnt/influxdb/data/blueshift/autogen/2/000000637-000000005.tsm.tmp (#0) 
> engine=tsm1
> Jul 28 20:40:29 ip-172-31-2-161 influxd[21528]: [I] 2017-07-28T20:40:29Z 
> compacted full 2 files into 1 files in 22.466184429s engine=tsm1
> Jul 28 21:20:40 ip-172-31-2-161 kernel: [242909.768230] 
> blk_update_request: I/O error, dev nvme0n1, sector 435160504
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: unexpected fault address 
> 0x7fa963c0a092
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: fatal error: fault
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: [signal SIGBUS: bus error 
> code=0x2 addr=0x7fa963c0a092 pc=0x9e2014]
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: goroutine 38 [running]:
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: runtime.throw(0xb82f57, 
> 0x5)
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> #011/usr/local/go/src/runtime/panic.go:596 +0x95 fp=0xc4203dfce8 
> sp=0xc4203dfcc8
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: runtime.sigpanic()
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> #011/usr/local/go/src/runtime/signal_unix.go:287 +0xf4 fp=0xc4203dfd38 
> sp=0xc4203dfce8
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> github.com/influxdata/influxdb/tsdb/engine/tsm1.(*indirectIndex).UnmarshalBinary(0xc4203ce280,
>  
> 0x7fa96272ae26, 0x58d4743, 0x58d474b, 0x0, 0x0)
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: #011/root/go/src/
> github.com/influxdata/influxdb/tsdb/engine/tsm1/reader.go:955 +0x1b4 
> fp=0xc4203dfdb0 sp=0xc4203dfd38
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> github.com/influxdata/influxdb/tsdb/engine/tsm1.(*mmapAccessor).init(0xc420463000,
>  
> 0x0, 0x0, 0x0)
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: #011/root/go/src/
> github.com/influxdata/influxdb/tsdb/engine/tsm1/reader.go:1053 +0x2d9 
> fp=0xc4203dfe30 sp=0xc4203dfdb0
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> github.com/influxdata/influxdb/tsdb/engine/tsm1.NewTSMReader(0xc4204fe098, 
> 0x27de2649, 0xef1580, 0x0)
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: #011/root/go/src/
> github.com/influxdata/influxdb/tsdb/engine/tsm1/reader.go:201 +0x136 
> fp=0xc4203dfea0 sp=0xc4203dfe30
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: 
> github.com/influxdata/influxdb/tsdb/engine/tsm1.(*FileStore).Open.func1(0xc42036e180,
>  
> 0xc4204f86c0, 0x3, 0xc4204fe098)
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: #011/root/go/src/
> github.com/influxdata/influxdb/tsdb/engine/tsm1/file_store.go:394 +0x63 
> fp=0xc4203dffc0 sp=0xc4203dfea0
> Jul 28 21:20:40 ip-172-31-2-161 influxd[23411]: runtime.goexit()
>
> Thank you
> Vinesh
>
>

-- 
Remember to include the version number!
--- 
You received this message because you are subscribed to the Google Groups 
"InfluxData" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/influxdb.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/influxdb/49bdea87-91d8-4066-be6d-96dd495e9336%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to