<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>PHP is not a good filesystem user. I've written about this a while back: <a href="https://joejulian.name/post/optimizing-web-performance-with-glusterfs/">https://joejulian.name/post/optimizing-web-performance-with-glusterfs/</a><br><br><div class="gmail_quote">On December 14, 2022 6:16:54 AM PST, Jaco Kroon <jaco@uls.co.za> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<p>Hi Peter,</p>
<p>Yes, we could. but with ~1000 vhosts that gets extremely
cumbersome to maintain and get clients to be able to manage their
own stuff. Essentially except if the htdocs/ folder is on a
single filesystem we're going to need to get involved with each
and every update, which isn't feasible. Then I'd rather partition
the vhosts such that half runs on one server and the other half on
the other server and risk downtime.</p>
<p>Our experience indicates that the slow part is in fact not the
execution of the php code but for php to locate the files. It
tries a bunch of folders with stat() and/or open() and gets the
ordering wrong, resulting numerous ENOENT errors before hitting
the right locations, after which it actually does quite well. On
code I wrote which does NOT suffer this problem quite as badly as
wordpress we find that from a local filesystem we get 200ms on
full processing (idle system, nvme physical disk, although I doubt
this matters since the fs layer should have most of this cached in
RAM anyway) vs 300ms on top of glusterfs. The bricks barely ever
goes to disk (fs layer caching) according to the system stats we
gathered.<br>
</p>
<p>How does big hosting entities like wordpress.org (iirc) deal with
this? Because honestly, I doubt they do single-server setups.
Then again, I reckon that if you ONLY host wordpress (based on
experience) it's possible to have a single master copy of
wordpress on each server, with a lsync'ed themes/ folder for each
vhost and a shared (glusterfs) uploads folder. Enters things like
wordfence that insists on being able to write to alternative
locations.<br>
</p>
<p>Anyway, barring using glusterfs we can certainly come up with
solutions, which may even include having *some* sites run on the
shared setup, and others on single-host, possibly with lsync
keeping a "semi hot standby" up to date with something like
lsync. That does get complex though.</p>
<p>Our ideal solution remains a fairly performant clustered
filesystem such as glusterfs (with which we have a lot of
experience, including using it for large email clusters where it's
performance is excellent, but I would have LOVED inotify
support). With nl-cache the performance is adequate, however, the
cache-invalidation doesn't seem to function properly. Which I
believe can be solved, either by fixing settings, or by fixing
code bugs. Basically whenver a file is modified or a new file is
created, clients should be alerted in order to invalidate cache.
Since this cluster is mostly-read, some write, and there is only
two clients, this should be perfectly manageable, and there seems
to be hints of this in the gluster volume options already:<br>
<br>
# gluster volume get volname all | grep invalid<br>
performance.quick-read-cache-invalidation false
(DEFAULT) <br>
performance.ctime-invalidation false
(DEFAULT) <br>
performance.cache-invalidation
on <br>
performance.global-cache-invalidation true
(DEFAULT) <br>
features.cache-invalidation
on <br>
features.cache-invalidation-timeout
600 <br>
<br>
</p>
<p>Kind Regards,<br>
Jaco</p>
<p> On 2022/12/14 14:56, Péter Károly JUHÁSZ wrote:<br>
</p>
<blockquote type="cite" cite="mid:CAAA01izvqKNdikAby07bjVja58_ogjjcSzT_=mYc5oWC=1ZEVA@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="auto">We did this with WordPress too. It uses a tons of
static files, executing them is the slow part. You can rsync
them and use the upload dir from glusterfs.</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">Jaco Kroon <<a href="mailto:jaco@uls.co.za" moz-do-not-send="true" class="moz-txt-link-freetext">jaco@uls.co.za</a>> 于
2022年12月14日周三 13:20写道：<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
<p>Hi,</p>
<p>The problem is files generated by wordpress, and uploads
etc ... so copying them to frontend hosts whilst making
perfect sense assumes I have control over the code to not
write to the local front-end, else we could have relied on
something like lsync.</p>
<p>As it stands, performance is acceptable with nl-cache
enabled, but the fact that we get those ENOENT errors are
highly problematic.<br>
</p>
<p><br>
</p>
<div>
<p>Kind Regards,<br>
Jaco Kroon<br>
</p>
<p><br>
</p>
<p>n 2022/12/14 14:04, Péter Károly JUHÁSZ wrote:<br>
</p>
</div>
<blockquote type="cite">
<div dir="auto">When we used glusterfs for websites, we
copied the web dir from gluster to local on frontend
boots, then served it from there.</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">Jaco Kroon <<a href="mailto:jaco@uls.co.za" target="_blank" rel="noreferrer" moz-do-not-send="true" class="moz-txt-link-freetext">jaco@uls.co.za</a>>
于 2022年12月14日周三 12:49写道：<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">Hi
All,<br>
<br>
We've got a glusterfs cluster that houses some php web
sites.<br>
<br>
This is generally considered a bad idea and we can see
why.<br>
<br>
With performance.nl-cache on it actually turns out to
be very <br>
reasonable, however, with this turned of performance
is roughly 5x <br>
worse. meaning a request that would take sub 500ms
now takes 2500ms. <br>
In other cases we see far, far worse cases, eg, with
nl-cache takes <br>
~1500ms, without takes ~30s (20x worse).<br>
<br>
So why not use nl-cache? Well, it results in readdir
reporting files <br>
which then fails to open with ENOENT. The cache also
never clears even <br>
though the configuration says nl-cache entries should
only be cached for <br>
60s. Even for "ls -lah" in affected folders you'll
notice ???? mark <br>
entries for attributes on files. If this recovers in
a reasonable time <br>
(say, a few seconds, sure).<br>
<br>
# gluster volume info<br>
Type: Replicate<br>
Volume ID: cbe08331-8b83-41ac-b56d-88ef30c0f5c7<br>
Status: Started<br>
Snapshot Count: 0<br>
Number of Bricks: 1 x 2 = 2<br>
Transport-type: tcp<br>
Options Reconfigured:<br>
performance.nl-cache: on<br>
cluster.readdir-optimize: on<br>
config.client-threads: 2<br>
config.brick-threads: 4<br>
config.global-threading: on<br>
performance.iot-pass-through: on<br>
storage.fips-mode-rchecksum: on<br>
cluster.granular-entry-heal: enable<br>
cluster.data-self-heal-algorithm: full<br>
cluster.locking-scheme: granular<br>
client.event-threads: 2<br>
server.event-threads: 2<br>
transport.address-family: inet<br>
nfs.disable: on<br>
cluster.metadata-self-heal: off<br>
cluster.entry-self-heal: off<br>
cluster.data-self-heal: off<br>
cluster.self-heal-daemon: on<br>
server.allow-insecure: on<br>
features.ctime: off<br>
performance.io-cache: on<br>
performance.cache-invalidation: on<br>
features.cache-invalidation: on<br>
performance.qr-cache-timeout: 600<br>
features.cache-invalidation-timeout: 600<br>
performance.io-cache-size: 128MB<br>
performance.cache-size: 128MB<br>
<br>
Are there any other recommendations short of abandon
all hope of <br>
redundancy and to revert to a single-server setup (for
the web code at <br>
least). Currently the cost of the redundancy seems to
outweigh the benefit.<br>
<br>
Glusterfs version 10.2. With patch for
--inode-table-size, mounts <br>
happen with:<br>
<br>
/usr/sbin/glusterfs --acl --reader-thread-count=2
--lru-limit=524288 <br>
--inode-table-size=524288 --invalidate-limit=16
--background-qlen=32 <br>
--fuse-mountopts=nodev,nosuid,noexec,noatime
--process-name fuse <br>
--volfile-server=127.0.0.1 --volfile-id=gv_home <br>
--fuse-mountopts=nodev,nosuid,noexec,noatime /home<br>
<br>
Kind Regards,<br>
Jaco<br>
<br>
________<br>
<br>
<br>
<br>
Community Meeting Calendar:<br>
<br>
Schedule -<br>
Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC<br>
Bridge: <a href="https://meet.google.com/cpu-eiue-hvk" rel="noreferrer noreferrer noreferrer" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">https://meet.google.com/cpu-eiue-hvk</a><br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" rel="noreferrer noreferrer" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">Gluster-users@gluster.org</a><br>
<a href="https://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer noreferrer noreferrer" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">https://lists.gluster.org/mailman/listinfo/gluster-users</a><br>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</blockquote></div><div class='k9mail-signature'>-- <br>Sent from my Android device with K-9 Mail. Please excuse my brevity.</div></body>
</html>