[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ceph-users
Subject:    [ceph-users] Re: High ceph_osd_commit_latency_ms on Toshiba MG07ACA14TE HDDs
From:       Lindsay Mathieson <lindsay.mathieson () gmail ! com>
Date:       2020-06-26 2:53:52
Message-ID: d092a4ce-2541-0f3b-e50e-b7b0de37e844 () gmail ! com
[Download RAW message or body]

On 25/06/2020 5:10 pm, Frank Schilder wrote:
> I was pondering with that. The problem is, that on Centos systems it seems to be ignored, in \
> general it does not apply to SAS drives, for example, and that it has no working way of \
> configuring which drives to exclude. 
> For example, while for data disks for ceph we have certain minimum requirements, like \
> functioning power loss protection, for an OS boot drive I really don't care. Power outages on \
> cheap drives that loose writes has not been a problem since ext4. A few log entries or \
> contents of swap - who cares. Here, performance is more important than data security on power \
> loss. 
> I would require a configurable option that works in the same way for all types of protocols, \
> SATA, SAS, NVMe disks, you name it. At time of writing, I don't know of any.


Yes, I can see that would be an issue for more upmarket systems to mine 
> ) Fortunately my cluster is small potatoes compared to most here, just 
34TB across 23 OSD's, all SATA. Given that, its easy enough to turn 
write caching of by default for my nodes and enable the OS drive via a 
startup script - I presume there are no cache flush issues when turning 
it on.


I did set this for the whole cluster, can't say I noticed any particular 
improvement in performance when testing from my VM's, but it certainly 
didn't degrade it either. And I felt it safe given the OSD safety issues 
mentioned earlier.


Everything is on a UPS, but nevertheless, stuff happens - turns out in 
our new office we share the switchboard with the office next door and 
the new load of our servers popped the circuit breakers overnight. So 
the neighbour took it upon himself to let himself in and turn our UPS 
off, taking our nodes down hard. No damage done fortunately, but words 
were spoken later.

-- 
Lindsay
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-leave@ceph.io


[prev in list] [next in list] [prev in thread] [next in thread]