One thing I'd love to see are benchmarks for ec profiles in different
host/network configurations. We're building a new cluster and I will
definitely be putting together an extensive benchmark panel together this
time around so that we know exactly what works best in what situations.
On Sat, Jun 2
If it creates it, should also remove it. Ceph-disk prepare did not
create it, so it is logical if it does not remove it. ceph-volume
however creates it, thus should remove it.
-Original Message-
From: David Turner [mailto:drakonst...@gmail.com]
Sent: zaterdag 2 juni 2018 23:36
To: M
Ceph-disk didn't remove an osd from the cluster either. That has never been
a thing for ceph-disk or ceph-volume. There are other commands for that.
On Sat, Jun 2, 2018, 4:29 PM Marc Roos wrote:
>
> But leaves still entries in crush map and maybe also ceph auth ls, and
> the dir in /var/lib/ceph
But leaves still entries in crush map and maybe also ceph auth ls, and
the dir in /var/lib/ceph/osd
-Original Message-
From: Oliver Freyermuth [mailto:freyerm...@physik.uni-bonn.de]
Sent: zaterdag 2 juni 2018 18:29
To: Marc Roos; ceph-users
Subject: Re: [ceph-users] Bug? ceph-volume
Am 02.06.2018 um 12:35 schrieb Marc Roos:
>
> o+w? I don’t think that is necessary not?
I also wondered about that, but it seems safe - it's only a tmpfs,
with sticky bit set - and all files within have:
-rw---.
as you can check.
Also, on our systems, we have:
drwxr-x---.
for /var/lib/ceph,
>>
>>
>> ceph-disk does not require bootstrap-osd/ceph.keyring and ceph-volume
>> does
>
>I believe that's expected when you use "prepare".
>For ceph-volume, "prepare" already bootstraps the OSD and fetches a
fresh OSD id, for which it needs the keyring.
>For ceph-disk, this was not par
Am 02.06.2018 um 11:44 schrieb Marc Roos:
>
>
> ceph-disk does not require bootstrap-osd/ceph.keyring and ceph-volume
> does
I believe that's expected when you use "prepare".
For ceph-volume, "prepare" already bootstraps the OSD and fetches a fresh OSD
id,
for which it needs the keyring.
For
The command mapping from ceph-disk to ceph-volume is certainly not 1:1.
What we are ended up using is:
ceph-volume lvm zap /dev/sda --destroy
This takes care of destroying Pvs and Lvs (as the documentation says).
Cheers,
Oliver
Am 02.06.2018 um 12:16 schrieb Marc Roos:
>
> I guess zap
dmcrypt is part of the whole device-mapper infrastructure :
https://en.wikipedia.org/wiki/Device_mapper
LVM is nothing but a tool to manipulate device-mapper easily
I do not see any reason for using dmcrypt directly without LVM
On 06/02/2018 11:24 AM, Marc Roos wrote:
>
>>> I would like to try
o+w? I don’t think that is necessary not?
drwxr-xr-x 2 ceph ceph 182 May 9 12:59 ceph-15
drwxr-xr-x 2 ceph ceph 182 May 9 20:51 ceph-14
drwxr-xr-x 2 ceph ceph 182 May 12 10:32 ceph-16
drwxr-xr-x 2 ceph ceph 6 Jun 2 17:21 ceph-19
drwxr-x--- 13 ceph ceph 168 Jun 2 17:47 .
drwxrwxrwt 2 ce
After removing manually the lv, vg, and pv, ceph-volume zap works, but
again the osd auth id still exists. (Although I have no idea how it
should know which to remove)
[@ bootstrap-osd]# ceph-volume lvm zap /dev/sdf
--> Zapping: /dev/sdf
Running command: /usr/sbin/cryptsetup status /dev/m
I guess zap should be used instead of destroy? Maybe keep ceph-disk
backwards compatibility and keep destroy??
[root@c03 bootstrap-osd]# ceph-volume lvm zap /dev/sdf
--> Zapping: /dev/sdf
--> Unmounting /var/lib/ceph/osd/ceph-19
Running command: umount -v /var/lib/ceph/osd/ceph-19
stderr: umou
> Kill all mds first , create new fs with old pools , then run ‘fs reset’
> before start any MDS.
Brilliant! I can't wait to try it.
Thanks.
--
Bryan Henderson San Jose, California
___
ceph-users mailing list
ceph-use
[@ bootstrap-osd]# ceph-volume lvm prepare --bluestore --data /dev/sdf
Running command: /bin/ceph-authtool --gen-print-key
Running command: /bin/ceph --cluster ceph --name client.bootstrap-osd
--keyring /var/lib/ceph/bootstrap-osd/ceph.keyring -i - osd new
c32036fe-ca0b-47d1-be3f-e28943ee3a97
ceph-disk does not require bootstrap-osd/ceph.keyring and ceph-volume
does
[@~]# ceph-disk prepare --bluestore --zap-disk /dev/sdf
***
Found invalid GPT and valid MBR; converting MBR to GPT format.
*
>> I would like to try this disk encryption without lvm.
>
>> And then have ceph use this device dmcrypt device
>
>You know that dmcrypt and LVM are tightly linked, right ?
No I did not. What makes this tightly linked relevant?
___
ceph-users mailing
On 06/02/2018 11:09 AM, Marc Roos wrote:
> I would like to try this disk encryption without lvm.
> And then have ceph use this device dmcrypt device
You know that dmcrypt and LVM are tightly linked, right ?
The later is a tool to handle the former
___
Is this documented somewhere? Because I would like to try this disk
encryption without lvm.
I thought about just creating a regular bluestore disk, and then use
dmcrypt on the 'ceph block' partition and use crypttab to automatically
mount it. And then have ceph use this device dmcrypt device
It would be nicer to keep such things on the mailinglist for future
reference external links expire etc.
-Original Message-
From: Vasu Kulkarni [mailto:vakul...@redhat.com]
Sent: vrijdag 1 juni 2018 18:51
To: ceph-users
Subject: Re: [ceph-users] Ceph EC profile, how are you using?
19 matches
Mail list logo