From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <pbs-devel-bounces@lists.proxmox.com>
Received: from firstgate.proxmox.com (firstgate.proxmox.com [IPv6:2a01:7e0:0:424::9])
by lore.proxmox.com (Postfix) with ESMTPS id D8A4B1FF16E
for <inbox@lore.proxmox.com>; Mon, 6 Jan 2025 20:08:11 +0100 (CET)
Received: from firstgate.proxmox.com (localhost [127.0.0.1])
by firstgate.proxmox.com (Proxmox) with ESMTP id 326D91D369;
Mon, 6 Jan 2025 20:08:00 +0100 (CET)
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=tuxis.nl; s=mail;
h=from:reply-to:subject:date:message-id:to:cc:mime-version:content-type:
in-reply-to:references;
bh=cRqiXe2kc2Jom0e6E/YotBYlr+5MsS5qtL/b375U/IE=;
b=EyMhnDrGHSGMbYmRQfPZl7EAEdFaPTv+yjeOvjW/uJVtPXtV2TbouOisUqokkP38kYbtOPOwQay2w
7BS0AjmvWXDbSVR2ptjkGFaryEIYBuCRBpCYeHqe56UQu4EdIj/PKROBt9dqIZ0F11To74ZTZyul4l
tMSUEHbHM7QG92o4ntVUBivCm3CBN0bxOnByV57siPI4IpU4/T3MbgX8WN5vpIB+dAly6QsLGAZLXv
IACtdIRRF/RlD1fpKpEwSiRE4NL/mHHHos4Av1csNfJM1WpbpmlGLoJEHisHifILmT44KT9UnDbfYu
67urATywC64AKjrooEM+sJbLr2nzbkw==
X-Footer: dHV4aXMubmw=
From: "Mark Schouten" <mark@tuxis.nl>
To: "Shannon Sterz" <s.sterz@proxmox.com>
Date: Mon, 06 Jan 2025 19:07:43 +0000
Message-Id: <em808af812-c25f-4569-8e1c-231663ec413f@d1b4cc0c.com>
In-Reply-To: <D6GK5TIN2LD4.1AXEF95IHZITS@proxmox.com>
References: <ema6032dba-8585-4377-bec1-11a37159087c@192dede7.com>
<D6D024F1JMI5.QDFXDKCQMUCJ@proxmox.com>
<embeb48874-d400-4e69-ae0f-2cc56a39d592@93f95f61.com>
<D6D3QC6Y5H4S.1QHYHPHXK6RVR@proxmox.com>
<em2d4327d4-4607-458e-880d-9b6bfba58f0a@ef757d37.com>
<D6GK5TIN2LD4.1AXEF95IHZITS@proxmox.com>
User-Agent: eM_Client/10.1.4828.0
MIME-Version: 1.0
X-SPAM-LEVEL: Spam detection results: 0
AWL -0.108 Adjusted score from AWL reputation of From: address
BAYES_00 -1.9 Bayes spam probability is 0 to 1%
DKIM_SIGNED 0.1 Message has a DKIM or DK signature,
not necessarily valid
DKIM_VALID -0.1 Message has at least one valid DKIM or DK signature
DKIM_VALID_AU -0.1 Message has a valid DKIM or DK signature from author's
domain
DKIM_VALID_EF -0.1 Message has a valid DKIM or DK signature from envelope-from
domain DMARC_PASS -0.1 DMARC pass policy
HTML_MESSAGE 0.001 HTML included in message
KAM_LOTSOFHASH 0.25 Emails with lots of hash-like gibberish
RCVD_IN_VALIDITY_CERTIFIED_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to
Validity was blocked. See
https://knowledge.validity.com/hc/en-us/articles/20961730681243 for more
information.
RCVD_IN_VALIDITY_RPBL_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to
Validity was blocked. See
https://knowledge.validity.com/hc/en-us/articles/20961730681243 for more
information.
RCVD_IN_VALIDITY_SAFE_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to
Validity was blocked. See
https://knowledge.validity.com/hc/en-us/articles/20961730681243 for more
information.
SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record
SPF_PASS -0.001 SPF: sender matches SPF record
Subject: Re: [pbs-devel] Authentication performance
X-BeenThere: pbs-devel@lists.proxmox.com
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Proxmox Backup Server development discussion
<pbs-devel.lists.proxmox.com>
List-Unsubscribe: <https://lists.proxmox.com/cgi-bin/mailman/options/pbs-devel>,
<mailto:pbs-devel-request@lists.proxmox.com?subject=unsubscribe>
List-Archive: <http://lists.proxmox.com/pipermail/pbs-devel/>
List-Post: <mailto:pbs-devel@lists.proxmox.com>
List-Help: <mailto:pbs-devel-request@lists.proxmox.com?subject=help>
List-Subscribe: <https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel>,
<mailto:pbs-devel-request@lists.proxmox.com?subject=subscribe>
Reply-To: Mark Schouten <mark@tuxis.nl>,
Proxmox Backup Server development discussion
<pbs-devel@lists.proxmox.com>
Cc: Proxmox Backup Server development discussion <pbs-devel@lists.proxmox.com>
Content-Type: multipart/mixed; boundary="===============2381593319938800892=="
Errors-To: pbs-devel-bounces@lists.proxmox.com
Sender: "pbs-devel" <pbs-devel-bounces@lists.proxmox.com>
--===============2381593319938800892==
Content-Type: multipart/alternative;
boundary="------=_MB1AC0B637-F58C-475C-9DC1-7B1A6ACD3A66"
--------=_MB1AC0B637-F58C-475C-9DC1-7B1A6ACD3A66
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: quoted-printable
https://bugzilla.proxmox.com/show_bug.cgi?id=3D6049 has been created for=20
this.
Thanks!
=E2=80=94
Mark Schouten
CTO, Tuxis B.V.
+31 318 200208 / mark@tuxis.nl
------ Original Message ------
>From "Shannon Sterz" <s.sterz@proxmox.com>
To "Mark Schouten" <mark@tuxis.nl>
Cc "Proxmox Backup Server development discussion"=20
<pbs-devel@lists.proxmox.com>
Date 20/12/2024 14:22:18
Subject Re: Re[4]: [pbs-devel] Authentication performance
>On Thu Dec 19, 2024 at 10:56 AM CET, Mark Schouten wrote:
>> Hi,
>>
>> We upgraded to 3.3 yesterday, not much gain to notice with regards to
>> the new version or the change in keying. It=E2=80=99s still (obvioulsy) =
pretty
>> busy.
>
>just be aware that the patch i linked to in my last mail has not been
>packaged yet, so you wouldn't see the impact of that patch yet.
>
>> However, I also tried to remove some datastores, which failed with
>> timeouts. PBS even stopped authenticating (so probably just working) al=
l
>> together for about 10 seconds, which was an unpleasant surprise.
>>
>> So looking into that further, I noticed the following logging:
>> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[39143]: GET
>> /api2/json/admin/datastore/XXXXXX/status: 400 Bad Request: [client
>> [::ffff]:42104] Unable to acquire lock
>> "/etc/proxmox-backup/.datastore.lck" - Interrupted system call (os erro=
r
>> 4)
>> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[39143]: GET
>> /api2/json/admin/datastore/XXXXXX/status: 400 Bad Request: [client
>> [::ffff]:42144] Unable to acquire lock
>> "/etc/proxmox-backup/.datastore.lck" - Interrupted system call (os erro=
r
>> 4)
>> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[39143]: GET
>> /api2/json/admin/datastore/XXXXXX/status: 400 Bad Request: [client
>> [::ffff]:47286] Unable to acquire lock
>> "/etc/proxmox-backup/.datastore.lck" - Interrupted system call (os erro=
r
>> 4)
>> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[39143]: GET
>> /api2/json/admin/datastore/XXXXXX/status: 400 Bad Request: [client
>> [::ffff:]:45994] Unable to acquire lock
>> "/etc/proxmox-backup/.datastore.lck" - Interrupted system call (os erro=
r
>> 4)
>>
>> Which surprised me, since this is a =E2=80=99status=E2=80=99 call, whic=
h should not need
>> locking of the datastore-config.
>>
>>https://git.proxmox.com/?p=3Dproxmox-backup.git;a=3Dblob;f=3Dsrc/api2/adm=
in/datastore.rs;h=3Dc611f593624977defc49d6e4de2ab8185cfe09e9;hb=3DHEAD#l687
>> does not lock the config, but
>>
>>https://git.proxmox.com/?p=3Dproxmox-backup.git;a=3Dblob;f=3Dpbs-datastor=
e/src/datastore.rs;h=3D0801b4bf6b25eaa6f306c7b39ae2cfe81b4782e1;hb=3DHEAD#l=
204
>> does.
>>
>> So if I understand this correctly, every =E2=80=99status=E2=80=99 call=
(30 per second in
>> our case) locks the datastore-config exclusively. And also, every time
>> =E2=80=99status=E2=80=99 get called, the whole datastore-config gets lo=
aded?
>
>probably, there are some comments about that there already, it might
>make sense to open a bugzilla issue to discuss this further [1].
>
>[1]: https://bugzilla.proxmox.com/
>
>> Is that something that could use some performance tuning?
>>
>> =E2=80=94
>> Mark Schouten
>> CTO, Tuxis B.V.
>> +31 318 200208 / mark@tuxis.nl
>>
>>
>> ------ Original Message ------
>> From "Shannon Sterz" <s.sterz@proxmox.com>
>> To "Mark Schouten" <mark@tuxis.nl>
>> Cc "Proxmox Backup Server development discussion"
>> <pbs-devel@lists.proxmox.com>
>> Date 16/12/2024 12:51:47
>> Subject Re: Re[2]: [pbs-devel] Authentication performance
>>
>> >On Mon Dec 16, 2024 at 12:23 PM CET, Mark Schouten wrote:
>> >> Hi,
>> >>
>> >> >
>> >> >would you mind sharing either `authkey.pub` or the output of the
>> >> >following commands:
>> >> >
>> >> >head --lines=3D1 /etc/proxmox-backup/authkey.key
>> >> >cat /etc/proxmox-backup/authkey.key | wc -l
>> >>
>> >> -----BEGIN RSA PRIVATE KEY-----
>> >> 51
>> >>
>> >> So that is indeed the legacy method. We are going to upgrade our PB=
S=E2=80=99es
>> >> on wednesday.
>> >>
>> >> >
>> >> >The first should give the PEM header of the authkey whereas the se=
cond
>> >> >provides the amount of lines that the key takes up in the file. Bo=
th
>> >> >give an indication whether you are using the legacy RSA keys or ne=
wer
>> >> >Ed25519 keys. The later should provide more performance, security=
should
>> >> >not be affected much by this change. If the output of the commands =
look
>> >> >like this:
>> >> >
>> >> >-----BEGIN PRIVATE KEY-----
>> >> >3
>> >> >
>> >> >Then you are using the newer keys. There currently isn't a recomme=
nded
>> >> >way to upgrade the keys. However, in theory you should be able to=
remove
>> >> >the old keys, re-start PBS and it should just generate keys in the =
new
>> >> >format. Note that this will logout anyone that is currently
>> >> >authenticated and they'll have to re-authenticate.
>> >>
>> >> Seems like a good moment to update those keys as well.
>> >
>> >Sure, just be aware that you have to manually delete the key before
>> >restarting the PBS. Upgrading alone won't affect the key. Ideally you'=
d
>> >test this before rolling it out, if you can
>> >
>> >> >In general, tokens should still be fater to authenticate so we'd
>> >> >recommend that you try to get your users to switch to token-based
>> >> >authentication where possible. Improving performance there is a bi=
t
>> >> >trickier though, as it often comes with a security trade-off (in t=
he
>> >> >background we use yescrypt fo the authentication there, that
>> >> >delibaretely adds a work factor). However, we may be able to impro=
ve
>> >> >performance a bit via caching methods or similar.
>> >>
>> >> Yes, that might help. I=E2=80=99m also not sure if it actually is
>> >> authentication, or if it is the datastore-call that the PVE-environ=
ments
>> >> call. As you can see in your support issue 3153557, it looks like s=
ome
>> >> requests loop through all datastores, before responding with a limi=
ted
>> >> set of datastores.
>> >
>> >I looked at that ticket and yes, that is probably unrelated to
>> >authentication.
>> >
>> >> For instance (and I=E2=80=99m a complete noob wrt Rust) but if I un=
derstand
>> >>https://git.proxmox.com/?p=3Dproxmox-backup.git;a=3Dblob;f=3Dsrc/api2=
/admin/datastore.rs;h=3D11d2641b9ca2d2c92da1a85e4cb16d780368abd3;hb=3DHEAD#=
l1315
>> >> correcly, PBS loops through all the datastores, checks mount-status =
and
>> >> config, and only starts filtering at line 1347. If I understand tha=
t
>> >> correctly, in our case with over 1100 datastores, that might cause=
quite
>> >> some load?
>> >
>> >Possible, yes, that would depend on your configuration. Are all of the=
se
>> >datastores defined with a backing device? Because if not, than this
>> >should be fairly fast (as in, this should not actually touch the disks=
).
>> >If they are, then yes this could be slow as each store would trigger a=
t
>> >least 2 stat calls afaict.
>> >
>> >In any case, it should be fine to move the `mount_status` check after
>> >the `if allowed || allow_id` check from what i can tell. Not sure why
>> >we'd need to check the mount_status for a datastore we won't include i=
n
>> >the resulsts anyway. Same goes for parsing the store config imo. Send=
a
>> >patch for that [1].
>> >
>> >[1]: https://lore.proxmox.com/pbs-devel/20241216115044.208595-1-s.ster=
z@proxmox.com/T/#u
>> >
>> >>
>> >>
>> >> Thanks,
>> >>
>> >> =E2=80=94
>> >> Mark Schouten
>> >> CTO, Tuxis B.V.
>> >> +31 318 200208 / mark@tuxis.nl
>> >
>> >
>
>
--------=_MB1AC0B637-F58C-475C-9DC1-7B1A6ACD3A66
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable
<html><head>
<style id=3D"css_styles" type=3D"text/css"><!--blockquote.cite { margin-lef=
t: 5px; margin-right: 0px; padding-left: 10px; padding-right:0px; border-le=
ft: 1px solid #cccccc }
blockquote.cite2 {margin-left: 5px; margin-right: 0px; padding-left: 10px;=
padding-right:0px; border-left: 1px solid #cccccc; margin-top: 3px; padding=
-top: 0px; }
a img { border: 0px; }
table { border-collapse: collapse; }
li[style=3D'text-align: center;'], li[style=3D'text-align: center; '], li[s=
tyle=3D'text-align: right;'], li[style=3D'text-align: right; '] { list-sty=
le-position: inside;}
body { font-family: Helvetica; font-size: 9pt; }
.quote { margin-left: 1em; margin-right: 1em; border-left: 5px #ebebeb soli=
d; padding-left: 0.3em; }
a.em-mention[href] { text-decoration: none; color: inherit; border-radius:=
3px; padding-left: 2px; padding-right: 2px; background-color: #e2e2e2; }
--></style></head>
<body style=3D"overflow-wrap: break-word; -webkit-nbsp-mode: space; line-br=
eak: after-white-space;"><div><a href=3D"https://bugzilla.proxmox.com/show_=
bug.cgi?id=3D6049">https://bugzilla.proxmox.com/show_bug.cgi?id=3D6049</a>=
=C2=A0has been created for this.</div><div><br /></div><div>Thanks!</div>
<div><br /></div><div id=3D"signature_old" style=3D"clear:both"><div style=
=3D"margin: 0px; padding: 0px; box-sizing: content-box;">=E2=80=94=C2=A0</d=
iv><div style=3D"margin: 0px; padding: 0px; box-sizing: content-box;">Mark=
Schouten</div><div style=3D"margin: 0px; padding: 0px; box-sizing: content-=
box;">CTO, Tuxis B.V.</div><div style=3D"margin: 0px; padding: 0px; box-siz=
ing: content-box;">+31 318 200208 /=C2=A0mark@tuxis.nl</div></div><div><br=
/></div>
<div x-em-replyforwardheader=3D""><br /></div>
<div>
<div>------ Original Message ------</div>
<div>From "Shannon Sterz" <<a href=3D"mailto:s.sterz@proxmox.com">s.ster=
z@proxmox.com</a>></div>
<div>To "Mark Schouten" <<a href=3D"mailto:mark@tuxis.nl">mark@tuxis.nl<=
/a>></div>
<div>Cc "Proxmox Backup Server development discussion" <<a href=3D"mailt=
o:pbs-devel@lists.proxmox.com">pbs-devel@lists.proxmox.com</a>></div>
<div>Date 20/12/2024 14:22:18</div>
<div>Subject Re: Re[4]: [pbs-devel] Authentication performance</div></div><=
div x-em-quote=3D""><br /></div>
<div id=3D"xd7a8ca7f59be40f" class=3D"plain"><blockquote cite=3D"D6GK5TIN2L=
D4.1AXEF95IHZITS@proxmox.com" type=3D"cite" class=3D"cite2">
<div class=3D"plain_line">On Thu Dec 19, 2024 at 10:56 AM CET, Mark Schoute=
n wrote:</div>
<blockquote type=3D"cite" class=3D"cite">
<div class=3D"plain_line"> Hi,</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> We upgraded to 3.3 yesterday, not much gain to n=
otice with regards to</div>
<div class=3D"plain_line"> the new version or the change in keying. It=E2=
=80=99s still (obvioulsy) pretty</div>
<div class=3D"plain_line"> busy.</div>
</blockquote>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line">just be aware that the patch i linked to in my la=
st mail has not been</div>
<div class=3D"plain_line">packaged yet, so you wouldn't see the impact of t=
hat patch yet.</div>
<div class=3D"plain_line">=C2=A0</div>
<blockquote type=3D"cite" class=3D"cite2">
<div class=3D"plain_line"> However, I also tried to remove some datastores, =
which failed with</div>
<div class=3D"plain_line"> timeouts. PBS even stopped authenticating (so pr=
obably just working) all</div>
<div class=3D"plain_line"> together for about 10 seconds, which was an unpl=
easant surprise.</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> So looking into that further, I noticed the foll=
owing logging:</div>
<div class=3D"plain_line"> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[3914=
3]: GET</div>
<div class=3D"plain_line"> /api2/json/admin/datastore/XXXXXX/status: 400 Ba=
d Request: [client</div>
<div class=3D"plain_line"> [::ffff]:42104] Unable to acquire lock</div>
<div class=3D"plain_line"> "/etc/proxmox-backup/.datastore.lck" - Interrupt=
ed system call (os error</div>
<div class=3D"plain_line"> 4)</div>
<div class=3D"plain_line"> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[3914=
3]: GET</div>
<div class=3D"plain_line"> /api2/json/admin/datastore/XXXXXX/status: 400 Ba=
d Request: [client</div>
<div class=3D"plain_line"> [::ffff]:42144] Unable to acquire lock</div>
<div class=3D"plain_line"> "/etc/proxmox-backup/.datastore.lck" - Interrupt=
ed system call (os error</div>
<div class=3D"plain_line"> 4)</div>
<div class=3D"plain_line"> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[3914=
3]: GET</div>
<div class=3D"plain_line"> /api2/json/admin/datastore/XXXXXX/status: 400 Ba=
d Request: [client</div>
<div class=3D"plain_line"> [::ffff]:47286] Unable to acquire lock</div>
<div class=3D"plain_line"> "/etc/proxmox-backup/.datastore.lck" - Interrupt=
ed system call (os error</div>
<div class=3D"plain_line"> 4)</div>
<div class=3D"plain_line"> Dec 18 16:14:32 pbs005 proxmox-backup-proxy[3914=
3]: GET</div>
<div class=3D"plain_line"> /api2/json/admin/datastore/XXXXXX/status: 400 Ba=
d Request: [client</div>
<div class=3D"plain_line"> [::ffff:]:45994] Unable to acquire lock</div>
<div class=3D"plain_line"> "/etc/proxmox-backup/.datastore.lck" - Interrupt=
ed system call (os error</div>
<div class=3D"plain_line"> 4)</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> Which surprised me, since this is a =E2=80=99sta=
tus=E2=80=99 call, which should not need</div>
<div class=3D"plain_line"> locking of the datastore-config.</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> <a href=3D"https://git.proxmox.com/?p=3Dproxmox-=
backup.git;a=3Dblob;f=3Dsrc/api2/admin/datastore.rs;h=3Dc611f593624977defc4=
9d6e4de2ab8185cfe09e9;hb=3DHEAD#l687" class=3D"__cef_visited">https://git.p=
roxmox.com/?p=3Dproxmox-backup.git;a=3Dblob;f=3Dsrc/api2/admin/datastore.rs=
;h=3Dc611f593624977defc49d6e4de2ab8185cfe09e9;hb=3DHEAD#l687</a></div>
<div class=3D"plain_line"> does not lock the config, but</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> <a href=3D"https://git.proxmox.com/?p=3Dproxmox-=
backup.git;a=3Dblob;f=3Dpbs-datastore/src/datastore.rs;h=3D0801b4bf6b25eaa6=
f306c7b39ae2cfe81b4782e1;hb=3DHEAD#l204" class=3D"__cef_visited">https://gi=
t.proxmox.com/?p=3Dproxmox-backup.git;a=3Dblob;f=3Dpbs-datastore/src/datast=
ore.rs;h=3D0801b4bf6b25eaa6f306c7b39ae2cfe81b4782e1;hb=3DHEAD#l204</a></div=
>
<div class=3D"plain_line"> does.</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> So if I understand this correctly, every =E2=80=
=99status=E2=80=99 call (30 per second in</div>
<div class=3D"plain_line"> our case) locks the datastore-config exclusively=
. And also, every time</div>
<div class=3D"plain_line"> =E2=80=99status=E2=80=99 get called, the whole d=
atastore-config gets loaded?</div>
</blockquote>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line">probably, there are some comments about that ther=
e already, it might</div>
<div class=3D"plain_line">make sense to open a bugzilla issue to discuss th=
is further [1].</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line">[1]: <a href=3D"https://bugzilla.proxmox.com/">ht=
tps://bugzilla.proxmox.com/</a></div>
<div class=3D"plain_line">=C2=A0</div>
<blockquote type=3D"cite" class=3D"cite2">
<div class=3D"plain_line"> Is that something that could use some performanc=
e tuning?</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> =E2=80=94</div>
<div class=3D"plain_line"> Mark Schouten</div>
<div class=3D"plain_line"> CTO, Tuxis B.V.</div>
<div class=3D"plain_line"> +31 318 200208 / <a href=3D"mailto:mark@tuxis.nl=
">mark@tuxis.nl</a></div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> ------ Original Message ------</div>
<div class=3D"plain_line"> From "Shannon Sterz" <<a href=3D"mailto:s.ste=
rz@proxmox.com">s.sterz@proxmox.com</a>></div>
<div class=3D"plain_line"> To "Mark Schouten" <<a href=3D"mailto:mark@tu=
xis.nl">mark@tuxis.nl</a>></div>
<div class=3D"plain_line"> Cc "Proxmox Backup Server development discussion=
"</div>
<div class=3D"plain_line"> <<a href=3D"mailto:pbs-devel@lists.proxmox.co=
m">pbs-devel@lists.proxmox.com</a>></div>
<div class=3D"plain_line"> Date 16/12/2024 12:51:47</div>
<div class=3D"plain_line"> Subject Re: Re[2]: [pbs-devel] Authentication pe=
rformance</div>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line"> >On Mon Dec 16, 2024 at 12:23 PM CET, Mark Sc=
houten wrote:</div>
<div class=3D"plain_line"> >> Hi,</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> ></div>
<div class=3D"plain_line"> >> >would you mind sharing either `aut=
hkey.pub` or the output of the</div>
<div class=3D"plain_line"> >> >following commands:</div>
<div class=3D"plain_line"> >> ></div>
<div class=3D"plain_line"> >> >head --lines=3D1 /etc/proxmox-back=
up/authkey.key</div>
<div class=3D"plain_line"> >> >cat /etc/proxmox-backup/authkey.ke=
y | wc -l</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> -----BEGIN RSA PRIVATE KEY-----</div>
<div class=3D"plain_line"> >> 51</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> So that is indeed the legacy method. W=
e are going to upgrade our PBS=E2=80=99es</div>
<div class=3D"plain_line"> >> on wednesday.</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> ></div>
<div class=3D"plain_line"> >> >The first should give the PEM head=
er of the authkey whereas the second</div>
<div class=3D"plain_line"> >> >provides the amount of lines that=
the key takes up in the file. Both</div>
<div class=3D"plain_line"> >> >give an indication whether you are =
using the legacy RSA keys or newer</div>
<div class=3D"plain_line"> >> >Ed25519 keys. The later should pro=
vide more performance, security should</div>
<div class=3D"plain_line"> >> >not be affected much by this chang=
e. If the output of the commands look</div>
<div class=3D"plain_line"> >> >like this:</div>
<div class=3D"plain_line"> >> ></div>
<div class=3D"plain_line"> >> >-----BEGIN PRIVATE KEY-----</div>
<div class=3D"plain_line"> >> >3</div>
<div class=3D"plain_line"> >> ></div>
<div class=3D"plain_line"> >> >Then you are using the newer keys. =
There currently isn't a recommended</div>
<div class=3D"plain_line"> >> >way to upgrade the keys. However,=
in theory you should be able to remove</div>
<div class=3D"plain_line"> >> >the old keys, re-start PBS and it=
should just generate keys in the new</div>
<div class=3D"plain_line"> >> >format. Note that this will logout =
anyone that is currently</div>
<div class=3D"plain_line"> >> >authenticated and they'll have to=
re-authenticate.</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> Seems like a good moment to update t=
hose keys as well.</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >Sure, just be aware that you have to manuall=
y delete the key before</div>
<div class=3D"plain_line"> >restarting the PBS. Upgrading alone won't af=
fect the key. Ideally you'd</div>
<div class=3D"plain_line"> >test this before rolling it out, if you can<=
/div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >> >In general, tokens should still be =
fater to authenticate so we'd</div>
<div class=3D"plain_line"> >> >recommend that you try to get your =
users to switch to token-based</div>
<div class=3D"plain_line"> >> >authentication where possible. Imp=
roving performance there is a bit</div>
<div class=3D"plain_line"> >> >trickier though, as it often comes =
with a security trade-off (in the</div>
<div class=3D"plain_line"> >> >background we use yescrypt fo the=
authentication there, that</div>
<div class=3D"plain_line"> >> >delibaretely adds a work factor).=
However, we may be able to improve</div>
<div class=3D"plain_line"> >> >performance a bit via caching meth=
ods or similar.</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> Yes, that might help. I=E2=80=99m also =
not sure if it actually is</div>
<div class=3D"plain_line"> >> authentication, or if it is the datast=
ore-call that the PVE-environments</div>
<div class=3D"plain_line"> >> call. As you can see in your support i=
ssue 3153557, it looks like some</div>
<div class=3D"plain_line"> >> requests loop through all datastores,=
before responding with a limited</div>
<div class=3D"plain_line"> >> set of datastores.</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >I looked at that ticket and yes, that is pro=
bably unrelated to</div>
<div class=3D"plain_line"> >authentication.</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >> For instance (and I=E2=80=99m a comple=
te noob wrt Rust) but if I understand</div>
<div class=3D"plain_line"> >><a href=3D"https://git.proxmox.com/?p=3D=
proxmox-backup.git;a=3Dblob;f=3Dsrc/api2/admin/datastore.rs;h=3D11d2641b9ca=
2d2c92da1a85e4cb16d780368abd3;hb=3DHEAD#l1315">https://git.proxmox.com/?p=
=3Dproxmox-backup.git;a=3Dblob;f=3Dsrc/api2/admin/datastore.rs;h=3D11d2641b=
9ca2d2c92da1a85e4cb16d780368abd3;hb=3DHEAD#l1315</a></div>
<div class=3D"plain_line"> >> correcly, PBS loops through all the da=
tastores, checks mount-status and</div>
<div class=3D"plain_line"> >> config, and only starts filtering at l=
ine 1347. If I understand that</div>
<div class=3D"plain_line"> >> correctly, in our case with over 1100=
datastores, that might cause quite</div>
<div class=3D"plain_line"> >> some load?</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >Possible, yes, that would depend on your con=
figuration. Are all of these</div>
<div class=3D"plain_line"> >datastores defined with a backing device? Be=
cause if not, than this</div>
<div class=3D"plain_line"> >should be fairly fast (as in, this should no=
t actually touch the disks).</div>
<div class=3D"plain_line"> >If they are, then yes this could be slow as=
each store would trigger at</div>
<div class=3D"plain_line"> >least 2 stat calls afaict.</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >In any case, it should be fine to move the `=
mount_status` check after</div>
<div class=3D"plain_line"> >the `if allowed || allow_id` check from what =
i can tell. Not sure why</div>
<div class=3D"plain_line"> >we'd need to check the mount_status for a da=
tastore we won't include in</div>
<div class=3D"plain_line"> >the resulsts anyway. Same goes for parsing t=
he store config imo. Send a</div>
<div class=3D"plain_line"> >patch for that [1].</div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >[1]: <a href=3D"https://lore.proxmox.com/pbs=
-devel/20241216115044.208595-1-s.sterz@proxmox.com/T/#u">https://lore.proxm=
ox.com/pbs-devel/20241216115044.208595-1-s.sterz@proxmox.com/T/#u</a></div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> Thanks,</div>
<div class=3D"plain_line"> >></div>
<div class=3D"plain_line"> >> =E2=80=94</div>
<div class=3D"plain_line"> >> Mark Schouten</div>
<div class=3D"plain_line"> >> CTO, Tuxis B.V.</div>
<div class=3D"plain_line"> >> +31 318 200208 / <a href=3D"mailto:mar=
k@tuxis.nl">mark@tuxis.nl</a></div>
<div class=3D"plain_line"> ></div>
<div class=3D"plain_line"> ></div>
</blockquote>
<div class=3D"plain_line">=C2=A0</div>
<div class=3D"plain_line">=C2=A0</div>
</blockquote></div>
</body></html>
--------=_MB1AC0B637-F58C-475C-9DC1-7B1A6ACD3A66--
--===============2381593319938800892==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
pbs-devel mailing list
pbs-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel
--===============2381593319938800892==--