Re: dovecot config for 1500 simultaneous connection

Hello,

On Mon, 13 Feb 2017 22:53:23 +0530 Rajesh M wrote:

thanks for your help

happy to say that the performance dramatically improved after i use the high performance settings from here http://wiki.dovecot.org/LoginProcess

That's why that page is there.

grep Login: /var/log/mail.log.1 |wc -l with the mail.log being of a typical, busy day. 412992

So about 5 login/s, not that busy, but that of course also depends on your HW and authentication backend.

i also picked up the imap and pop3 connections during peak hours [root@ns1 domains]# doveadm who | awk '/imap/{m+=$2}/pop3/{n+=$2}END{print m,n}' username # proto (pids) (ips)

max figures i got was 535 for imap and 97 for pop

Question 1 i wish to improve the performance further by caching the logins. current the same is kept disable because when user's change passwords then they are not able to immediately login with the new password for some time. How to solve this issue.

There is no solution, caching auth data is by default fraught with that sort of peril.

And how did you determine that caching auth data would actually improve things?

What is your authentication backend?

With 4000 users and 5 logins/s I can't imagine anything that wouldn't be able to: a) put all of the relevant data into RAM and b) thus handle auth requests in very timely manner.

The example I gave earlier about my servers with 140 logins/s authenticates against LDAP and isn't very busy doing so while also fitting all 2.7GB of data for about 1 million users into RAM.

further to the above. i am planning to put up a large mail server with one of the following options

Don't plan 1, always plan a pair and then use whatever you're comfortable with to turn them into a HA cluster, like Pacemaker with DRBD.

option 1 OS and mailqueue : 2 X 600 gb 15k rpm raid 1 What kind of RAID, HW or MD?

data : 4 X 2000 gb raid 10 for data backup : 2 X 2000 gb raid 10 for backup

While with Linux MD a 2 disk RAID10 would be possible, it wouldn't be particular helpful (slower than a RAID 1 with writes).

32 gb ram

option 2 os and mail queue : 4 X 600 gb 15k rpm raid 10 Those 15k HDDs tend to be expensive. Estimate your expected writes per day (iostat on an existing server can give you some hints) and get 2 SSDs instead, DC S3520s if 1 DWPD is sufficient, DC S3610s if you need 3 DWPD (doubtful), S3710 if you need 10 DWPD (very doubtful). Similar for Samsung DC level SSDs, check their specs.

For example, my busiest mailbox servers only write about 250KB/s averaged, which mean about 50GB/day and thus would be a fit even for a small S3520. I'm still using 200GB S3710s because I can afford them, like the speed and never ever want to worry about wearout.

data : 4 X 2000 gb raid 10 for data 32 gb ram

i am having the OS and mail queue on primary drives

i understand that Raid10 give a faster write and read whereas Raid 1 will have slow writes and fast reads

While that is true, you're missing the real difference here, the number of HDDs. With 4 of them in the RAID10 they will be twice as fast as the 2 disk RAID1. This can be tweaked even further with the various RAID 10 layout options, "man md".

the question is will there be huge performance difference between Raid1 and Raid 10 when it comes small files like mail queue ?

Again as explained above, double the IOPS, so yes.

Christian

thanks rajesh ----- Original Message ----- From: Christian Balzer [mailto:chibi@gol.com] To: dovecot@dovecot.org Cc: kevin@my.walr.us Sent: Mon, 13 Feb 2017 10:46:13 +0900 Subject:

Hello,

On Sun, 12 Feb 2017 08:27:21 -0500 KT Walrus wrote:

...

Thanks for the info. I do have one further question for you. On your servers that are currently handling 50k IMAP sessions, how many users does that correspond to? Since many users will have multiple IMAP sessions on multiple devices, I’d like to hear about some real-world numbers that could be used for budgeting a new project like mine.

Those servers would actually be the wrong ones to look at, as they are primarily accessed by the aforementioned broken client, so the numbers are somewhat skewed. However looking at other servers with a more "normal" user spread things aren't actually too different.

The average number of sessions per user tends to be 2. The clear majority (over 50%) only has one session open (people with well behaved and configured clients watching the INBOX mostly). Another 30% has 2 sessions open, again the typical state of this would be clients watching another mailbox besides INBOX, typically SENT. The rest has 3 and more sessions open.

The number of sessions could of course be drastically reduced if any client would support IMAP NOTIFY, alas none that I'm aware of do.

Lastly no more than 60% of the session seem to be in IDLE at any given time, so my comments about RAM and IMAP hibernation effectiveness stand.

...

Also, do you use Dovecot IMAP proxies in front of your backend servers? If so, how many IMAP sessions can one proxy server handle (assuming the proxy does authorization using MySQL running on a separate server)? And, could the proxy server be tuned to help in optimizing mostly IDLE backend sessions?

Yes to Dovecot Proxying, of course.

No idea about MySQL, with (Open)LDAP nothing is breaking a sweat at an average of 140 logins per second (IMAP and POP) on the 2 proxy servers. If you can fit your entire dataset into RAM it should be fine, my LDAP servers fall into that category and take about 10% of a slow (1.2GHz, 34%, power-save mode) core only to handle the that load (also 2 servers). And the rate of logins/s is what you need to worry about most and optimize for.

The proxies will of course have to do the shuffling of data and SSL en/de-coding, but again they're not particular busy with that.

The number of sessions comes into play when looking at the number of login processes on the proxies and their memory footprint. An IMAP login process on the proxies in performance mode with a client limit of 1000 will consume about 55MB at most. So assume at least 55KB RAM per session.

Read Timo's mail I linked to about IMAP hibernation, AFAIK nothing has happened to make proxies more supportive for this though.

Christian

...

...

On Feb 12, 2017, at 1:58 AM, Christian Balzer chibi@gol.com wrote:

Hello,

On Fri, 10 Feb 2017 14:50:03 -0500 KT Walrus wrote:

...

...

1. 256GB of real RAM, swap is for chums.

Are you sure that 100,000 IMAP sessions wouldn’t work well with SWAP, especially with fast SSD storage (which is a lot cheaper than RAM)?

I'm sure about tax and death, not much else.

But as a rule of thumb I'd avoid swapping out stuff on production servers, even if it were to SSDs. Incidentally the servers I'm talking about here have their OS and swap on Intel DC S3710s (200GB) and the actual storage on plenty of 1.6TB DC S3610s.

Relying on the kernel to make swap decisions is likely to result in much reduced performance even with fast SWAP when you're overcommitting things on that scale.

But read on.

...

Seems that these IMAP processes are long lived processes (idling most of the time) that don’t need that much of the contents of real memory available for much of the life of the process. I use a database proxy in front of MySQL (for my web apps) so that there can be a large number of TCP connections to the proxy where the frontend requests are queued for execution using a small number of backend connections.

Could Dovecot IMAP be re-written to be more efficient so it works more like MySQL (or other scalable data servers) that could handle a million or more IMAP sessions on a server with 32GBs or less of RAM? Those IMAP sessions aren’t doing much most of the time and shouldn’t really average 2MB of active data per session that needs to be resident in main memory at all times.

See IMAP hibernation: https://www.mail-archive.com/dovecot@dovecot.org/msg63429.html https://www.mail-archive.com/dovecot@dovecot.org/msg63429.html

I'm going to deploy/test this in production in about 2 months from now, but if you look at the link and the consequent changelog entries you'll see that it has certain shortcomings and bug fixes in pretty much each release after it was introduced.

But this is the correct way to tackle things, not SWAP.

Alas I'm not expecting miracles and if more than 20% of the IMAP sessions here will be hibernated at any given time I'd be pleasantly surprised.

Because between:

1. Finding a sensible imap_hibernate_timeout.

2. Having well behaved clients that keep idling instead of restarting the sequence (https://joshdata.wordpress.com/2014/08/09/how-bad-is-imap-idle/ https://joshdata.wordpress.com/2014/08/09/how-bad-is-imap-idle/)

3. Having lots of mobile clients who either get disconnected (invisible to Dovecot) or have aggressive IDLE timers to overcome carrier NAT timeouts (a large mobile carrier here times out idle TCP sessions after 2 minutes, forcing people to use 1 minute IDLE renewals, making 1. up there a nightmare).

4. Having really broken clients (don't ask, I can't tell) which open IMAP sessions, don't put them into IDLE and thus having them expire after 30 minutes.

the pool of eligible IDLE sessions isn't as big as it could be, in my case at least.

...

My mail server isn’t that large yet as I haven’t fully deployed Dovecot outside my own small group yet, but it would be nice if scaling Dovecot IMAP to millions of users wasn’t limited to 50,000 IMAP sessions on a server...

Scaling up is nice and desirable from a cost (rack space, HW) perspective, but the scalability of things OTHER than Dovecot as I pointed out plus that little detail of failure domains (do you really want half of your eggs in one basket?) argue for scaling out after a certain density.

I'm feeling my way there at this time, but expect more than 100k sessions per server to be tricky.

Lastly, when I asked about 500k sessions per server here not so long ago, ( http://www.dovecot.org/list/dovecot/2016-November/106284.html http://www.dovecot.org/list/dovecot/2016-November/106284.html ) Timo mentioned that he's not aware of anybody doing more than 50k per server, something I got licked already and definitely will go to 100k eventually.

Regards,

Christian

...

...

On Feb 10, 2017, at 11:07 AM, Christian Balzer chibi@gol.com wrote:

On Fri, 10 Feb 2017 07:59:52 -0500 KT Walrus wrote:

...

> 1500 IMAP sessions will eat up about 3GB alone.

Are you saying that Dovecot needs 2MB of physical memory per IMAP session?

That depends on the IMAP session, read the mailbox size and index size, etc. Some are significantly larger:

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1033864 mail 20 0 97600 67m 54m S 0 0.1 0:01.15 imap

But yes, as somebody who has mailbox servers with 55k+ session the average is around 1.6MB.

...

If I want to support a max 100,000 IMAP sessions per server, I should configure the server to have at least 200GBs of SWAP?

You will want:

...

2. Understanding how to tune Dovecot and more importantly the overall system to such a task (see that PID up there?).
3. Be willing to deal with stuff like top and ps taking ages to start/run and others like atop actually killing dovecot (performance wise, not literally) when doing their obviously flawed cleanup on exit. Some things clearly do NOT scale well.

My current goal is to have 100k capable servers that work well, 200k in a failover scenario, but that won't be particular enjoyable.

Christian

...

> On Feb 10, 2017, at 3:58 AM, Christian Balzer chibi@gol.com wrote: > > On Fri, 10 Feb 2017 01:13:20 +0530 Rajesh M wrote: >
>> hello >> >> could somebody with experience let me know the dovecot config file settings to handle around 1500 simultaneous connections over pop3 and 1500 connection over imap simultaneously. >>
> > Be very precise here, you expect to see 1500 as the result of > "doveadm who |grep pop3 |wc -l"? > > Because that implies an ungodly number of POP3 connects per second, given > the typically short duration of these. > > 1500 IMAP connections (note that frequently a client will have more than > the INBOX open and thus have more than one session and thus process on the > server) are a much easier proposition, provided they are of the typical > long lasting type. > > So can you put a number to your expected logins per second (both protocols)? >
>> my server >> >> server configuration >> hex core processor, 16 gb ram 1 X 600 gb 15 k rpm for main drive and 2 X 2000 >> gb hdd for data (No raid) >>
> No RAID and no other replication like DRBD? > Why would you even bother? > > How many users/mailboxes in total with what quota? > > 1500 IMAP sessions will eat up about 3GB alone. > You will want more memory, simply to keep all relevant SLAB bits (inodes, > dentries) in RAM. > > If you really have several hundreds logins/s, you're facing several > bottlenecks: > 1. Login processes themselves (easily fixed by high performance mode) > 2. Auth processes (that will depend on your backends, method mostly) > 3. Dovecot master process (spawning mail processes) > > The later is a single-threaded process, so it will benefit from a faster > CPU core. > It can be dramatically improved by enabling process re-usage, see: > http://wiki.dovecot.org/PerformanceTuning > > However that also means more memory usage. > > > > Christian >
>> >> thanks >> rajesh >>
> > [snip] > -- > Christian Balzer Network/Systems Engineer
> chibi@gol.com Global OnLine Japan/Rakuten Communications > http://www.gol.com/

-- Christian Balzer Network/Systems Engineer
chibi@gol.com mailto:chibi@gol.com mailto:chibi@gol.com> Global OnLine Japan/Rakuten Communications http://www.gol.com/ http://www.gol.com/ <http://www.gol.com/ http://www.gol.com/>

-- Christian Balzer Network/Systems Engineer
chibi@gol.com mailto:chibi@gol.com Global OnLine Japan/Rakuten Communications http://www.gol.com/ http://www.gol.com/

-- Christian Balzer Network/Systems Engineer
chibi@gol.com Global OnLine Japan/Rakuten Communications http://www.gol.com/