By: Harrison Fisk
When we ported this patch to our branch, we fixed this hang as well: https://code.launchpad.net/~percona-dev/percona-server/fb_changes_auto_lru_dump/+merge/46012
View ArticleBy: Mark Leith
Both implementations can block buffer pool operations for certain periods of time. Looking at the fix for the above mentioned bug by Laurynas...
View ArticleBy: Xaprb
Yes. In case of a server crash, if the LRU dump is on a DRBD or SAN along with the data, then a cold standby can be brought up quickly. This was one of the major use cases for the feature originally....
View ArticleBy: Laurynas
Mark - Yes, our implementation holds the mutex for every 4096KB of dump, which amounts to scanning half of million pages on the LRU list and 8GB of buffer pool dumped. We may adjust this amount up or...
View ArticleBy: George O. Lorch III
There was in fact a slight error in our Percona Server 5.1 implementation when an I/O error occurred during write. This has been corrected and should appear in our next 5.1 release. In our Percona...
View Article