I have been having issues with a node. It is somehow disconnected from the cluster. Looking at the cluster status i get disconnected. When I try to do a select from a table I get WSREP has not yet prepared node for application use
. The cluster size on the disconnected cluster is 0, the others are reporting a cluster size of 4. The wsrep_address
for the disconnected cluster has none, the working cluster report the correct list of ip’s. Looking at the disconnected node’s config file. It’s the same as the others.
Hi @danarashad
Can you send us the error log entries from the node that disconnected, along with the same time period from another member of the cluster that didn’t evict? Thanks
I think it is a space issue according to this message.
[InnoDB] posix_fallocate(): Failed to preallocate data for file ./#innodb_temp/temp_8.ibt, desired size 16384 bytes. Operating system error number 28. Check that the disk is not full or a disk quota exceeded. Make sure the file system supports this function.
The datadir is 1.8t
and the usage is 1.4t
.
When a server is joining the cluster where does the data get stored? In the /tmpdir
listed in the config file then pushed to datadir? When data is being replicated to servers on the cluster where is that data stored /tmpdir
then pushed to datadir?
Also when joining the cluster does all of the data get copied to the tmpdir
then moved to datadir
or is it streamed to tmpdir
then streamed to datadir
to not fillup both directories?