In preparing for a proof of concept with very large tables, I was given the impression that there’s an option to enable compression on sst. Redirecting
The docs read like I can use compressor and decompressor options under [sst]. The server accepts the options. But, wsrep_sst_method=xtrabackup-v2 returns a “file not found” error as soon as it attempts to perform a sync. It does no good to set the whole path.
What am I missing?
1 Like
as error states it is looking for “pigz” binaries
this is needed to perform compression in parallel threads
1 Like
Thanks for the reply. It exists.
which pigz
/usr/bin/pigz
This is using AWS’ generic ubuntu v20 ec2 instance.
This is the config (uncommented, it generates the sst error)
#compressor = ’pigz -p 4 --fast’
#decompressor = ’pigz -p 4 -d’
1 Like
I am not really sure why PXC does not see it.
Can you check regular Linux functionality?
is pigz executable? is /usr/bin in PATH for PXC process?
1 Like
Pigz is definitely executable. I can successfully compress/decompress a file.
Excellent question. How can I check the definition of PATH that the PXC process uses?
I know the whole joining process launches socat to sync files. In Ubuntu 20.x, socat is in /usr/bin. Pigz is in /usr/bin as well.
1 Like