onlyjob
> We can add to the mfsmaster.cfg new option "DAYS_TO_REMOVE_UNUSED_CHUNKSERVER" (or something like that) with default value set to 30 or something like that. Sounds like a good idea. Let's...
Libvirt should be convinced that MFS is a shared file system similar to the following: * https://github.com/libvirt/libvirt/commit/6dd2a2ae6386b1d51edcc9a434f56d7f9dc2cb35 # Ceph * https://github.com/libvirt/libvirt/commit/d163b940a73a0d6b8277a4bccef2b60936933cf0 # GPFS
IMHO this issue is trivial enough to submit a merge request without searching for insiders...
Probably not in the near future... Definitely not soon...
IMHO it is a bug: when chunkserver starts it should behave the same as during runtime yet here it acts differently. However this is not a simple problem as you...
LZ4 overhead is negligible and its speed is close to RAM-to-RAM copy: > LZ4 is a very fast lossless compression algorithm, providing compression speed > 500 MB/s per core, scalable...
> * compression of chunk data in chunkserver - client communication - well, the problem is, we see more and more MooseFS installations using 10Gb networks and no compression algorithms...
Btrfs supports compression. But on large file systems its performance degrade significantly over time due to fragmentation. Also Btrfs requires periodic balancing. Unfortunately Btrfs is not the best fit for...
This is definitely not fixed. I just had another spike of insanely high replication rate (12_000 replications per minute) on a chart from a moment when disk was marked as...