similar to: Cache details

I'm using dovecot in a multi server setup. Two directors forward pop3, imap and smtp to backend dovecots with their central storage on NFS. Yesterday I stumbled upon these log lines: 2019-11-18 20:17:51 lmtp(17274): Info: Connect from 172.22.1.6 2019-11-18 20:17:51 lmtp(17274): Panic: file smtp-client-connection.c: line 1142 (smtp_client_connection_established): assertion failed:

Hi, today we have exceprienced really major difficulties with our proxy backend servers. Everything started after enabling auth cache: auth_cache_size = 100M auth_cache_verify_password_with_worker = yes auth_cache_ttl = 1 hour auth_cache_negative_ttl = 1 hour Soon after we start receiving lots of calls from our customers about 'missing all e-mail messages'. This was in errors logs on

Timo Sirainen wrote: > - auth: Auth cache is now disabled passwd-file. It was unnecessary and > it broke %variables in extra fields. So settings like auth_cache_size = 65536 auth_cache_ttl = 86400 auth_cache_negative_ttl = 600 are ignored/useless, and passwd-file auth lookups always results in file read operation? Joseph Tam <jtam.home at gmail.com>

Hi, I've been running dovecot for many years, but now i've hit a strange problem. when retrieving imap there is sometimes in 1 of 4 imap requests a 4sec latency. * notes: - connected directly to the backends this latency disappears - removing a director from the loadbalancer(lvs) so i'm the only connected to the director this latency disappears too I would appreciate some feedback

Hi Guys, We have: auth_cache_size = 2048 #Kb auth_cache_ttl = 900 #life of the cache Seems to work well (though haven't tried looking at stats). Only problem is that every so often, one of our new users tries to log in quickly before the account is marked (in a db) as 'active'. They use the right password but the sql query fails because the active flag is not yet set. This

Hi, Trying to keep abusive/buggy IMAP clients at bay on a number of Dovecot proxy servers, I've reconfigured them to use "mail_max_userip_connections = 50" in the "protocol imap" section, followed by restarting Dovecot. Yet, I'm still seeing 160+ established connections from a single IP address for the same email account. Am I missing anything? # 2.2.27

Hi, please post your doveconf -n Sami > On 16 Apr 2018, at 20.17, azurit at pobox.sk wrote: > > Hi, > > today we have exceprienced really major difficulties with our proxy backend servers. Everything started after enabling auth cache: > auth_cache_size = 100M > auth_cache_verify_password_with_worker = yes > auth_cache_ttl = 1 hour > auth_cache_negative_ttl = 1 hour

hi centos 6 64 bit hex core processor with hyperthreading ie display shows 12 cores 16 gb ram 600 gb 15000 rpm drive we are having around 4000 users on a server i wish to allow 1500 pop3 and 1500 imap connections simultaneously. need help regarding the settings to handle the above imap-login, pop3-login imap pop3 service settings i recently i got an error imap-login: Error: read(imap)

Edit: When using win8, then logs started appear in dovecot.rawlog. But as win10 gives error in ssl level, before user authentication, then no debug logs will be written into user dovecot.rawlog folder. So how to debug this ssl issue? Aki Tuomi wrote: > https://wiki2.dovecot.org/Debugging/Rawlog > > can you try this to get rawlogs to find out what happens? > > Aki I'm bit

Hello, I'm migrating many accounts to a new server with vpopmail 5.4.33 and dovecot 2.0.11. I've already vpopmail 5.4.32 and dovecot 1.2.16 on others servers running without problems. With dovecot 2.0.11 my lastauth file is not updated. This file usually is update on any access (smtp, pop3, imap) with the client's IP, for every mailbox. Now it's updated only when a client

# 2.3.4.1 (f79e8e7e4): /etc/dovecot/dovecot.conf Hello, we're using mail on NFS and the indexes on localdisk, with a director setup. (18K+ users) now to get more performance i thought to use the ITERINDEX option. however this has the side-effect on some mailboxes that it doesn't show the subfolders under INBOX. they seem 'disappeared'. Removing the ITERINDEX option fixes this

Hi all, I have observed with my Dovecot setup that unique failed logins cause legitimate correct logins to be slowed. I am running two servers, each with two Dovecot instances, a Proxy with Director, and a backend Dovecot. I suspect that the backend instance is throttling connections from the same IP, and because I am running a Proxy, the backend will only see either of the two server IPs. I

Hi, mail_max_userip_connections is only enforced at the backend level. The setting has no effect on proxy. If you want to force the limit then you can only do it in the backend. Sami > On 9 Mar 2017, at 12.05, Adi Pircalabu <adi at ddns.com.au> wrote: > > Quick follow-up: updated the proxies to 2.2.28, but I still couldn't find a way to limit the inbound IMAP connections per

Hey everyone, I've got a weird error since I upgraded to the latest dovecot on my FreeBSD box: root at postfixjail /# doveadm quota recalc -u xxx at xxx.com doveadm(xxx at xxx.com): Error: dict-client: Commit failed: Deinit fish: 'doveadm quota recalc -u xxx@?' terminated by signal SIGSEGV (Address boundary error) root at postfixjail /# dovecot --version 2.2.25 (7be1766) root at

----- Original Message ----- From: Tony Morehen [mailto:tmorehen at ajmconsulting.ca] To: dovecot at dovecot.org Sent: Sat, 12 Sep 2015 17:57:27 -0400 Subject: Re: concerning dovecot settings for high volume server You may be running up against Linux system/user limits. Run $ cat /proc/sys/kernel/pid_max and $ ulimit -a That should give some insight into your problem. On 12/09/2015 2:53 PM,

Just applied Debian SSL updates, due to vulnerabilities reported: "Package : openssl CVE ID : CVE-2019-1547 CVE-2019-1549 CVE-2019-1563 Three security issues were discovered in OpenSSL: A timing attack against ECDSA, a padding oracle in PKCS7_dataDecode() and CMS_decrypt_set1_pkey() and it was discovered that a feature of the random number generator (RNG) intended to

Hi, since migration to Dovecot 2.2 I have troubles with lazy_expunge as soon as a user shares a folder. The user the folder is shared to cannot login anymore, dovecot logs Fatal: lazy_expunge: Unknown namespace: '.EXPUNGED/' - When no folder is shared lazy_expunge is working fine, deleted mails are moved to the expunged namespace and can be recovered without any problems. - Sharing

thanks very much urban. this was very helpful. i have around 12500 users spread over 3 independent servers each having around 4000+ users i am using qmailtoaster, vpopmail, spamassassin and dovecot. in future i am planning to consolidate all using a HA cluster. if it is ok with you could you kindly share some information about your email server configuration. if you do not wish to put it on the

Rajesh M wrote: > 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. Dovecot shouldn't be doing that. According to https://wiki2.dovecot.org/Authentication/Caching Data is

Hi! After upgrading from Debian wheezy with (self compiled) dovecot 2.2.15 to Debian jessie with (self compiled) 2.2.24, I observe the following segmentation fault in the logs: Jun 7 09:23:09 imap dovecot: imap(user at example.com): Error: read(<imap client>) failed: read(size=8003) failed: Connection reset by peer (uid=0, box=trash) Jun 7 09:23:09 imap dovecot: imap(user at