We recently removed index.shard.check_on_startup:fix from our settings and
haven't had this problem since. The guide says "Should shard consistency be
checked upon opening" but it appears to also affect replication. I'm not
going to say that that is wrong although it isn't what I want but I think
that the guide should be more explicit as to when the checking is done.
On Tuesday, 29 April 2014 15:50:05 UTC+2, Michael Salmon wrote:
>
> I am having trouble replicating a shard and I cannot see any possible
> reason for it. After 15 minutes I get a timeout in phase 2.
>
> The shard isn't that large about 60,000K, 5GB and 22 segments and the
> translog directories are empty.
> The computers in question are lightly loaded as is the network between
> them.
> Copying all the files in the shard from all 4 disks between the two
> computers with rsync takes about 40 seconds.
> I can't run checkIndex on the source machine as it can't handle shards
> that are spread over multiple disks but it runs quite happily on the files
> I copied with rsync although it took a bit over 12 minutes to run the check.
> I have ES 1.1.0 installed.
> I changed some settings but none of them seem to make much difference:
>
> "transient": {
> "logger": {
> "level": "TRACE"
> },
> "indices": {
> "store": {
> "throttle": {
> "type": "none"
> }
> },
> "recovery": {
> "translog_size": "256MB",
> "concurrent_streams": "16",
> "translog_ops": "10000",
> "max_bytes_per_sec": "250MB"
> }
> }
> }
>
> Does anyone have any tips on how I should proceed?
>
>
--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/bde6fb91-7b3c-42d7-8e31-7fdb7bd5555b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
