1

Topic: Webaccess is down

==== Required information ====
- iRedMail version:
- Store mail accounts in which backend (LDAP/MySQL/PGSQL):
- Linux/BSD distribution name and version:
- Related log if you're reporting an issue:
============ Required information ====
- iRedMail version: 0.9.0
- Store mail accounts in which backend (LDAP/MySQL/PGSQL): MySQL
- Linux/BSD distribution name and version: Ubuntu Server 14.04
- Related log if you're reporting an issue:
====

Hello,
The VM that is currently runnign iRedMail has 32GB ram and 8 cores. The webserver has it's own valid SSL wildcard certificate.
Although this configuration, if an email transaction occurs, the webclient fails to respond. The nginx webserver is fine, but it isn't delivering content at all, not even static. System load drops to a nearly 0.28% and stays there.

I just launched this and noticed it. At the moment I have 3 domains and a total of 17 accounts.

Bottom line: if I receive an email or send one, the webclient is down. If i try to connect thunderbird, it dies. If i connect gmail, it dies.

Some extra info:

# grep 'max_servers' /etc/amavis/conf.d/50-user
$max_servers = 4;
# grep 'smtp-amavis' /etc/postfix/master.cf
#  -o content_filter=smtp-amavis:[127.0.0.1]:10026
smtp-amavis unix -  -   -   -   4  smtp

Some ideas? Thanks!

----

Spider Email Archiver: On-Premises, lightweight email archiving software developed by iRedMail team. Supports Amazon S3 compatible storage and custom branding.

2

Re: Webaccess is down

Found what the problem was. I've changed the redirects in nginx from /mail to /0 and after any kind of action(send, receive, email client sycn) the system broke.

3

Re: Webaccess is down

Is problem solved?

By the way, if you have 8 CPU cores, you can set '$max_servers = 8;' in amavisd, and 'smtp-amavis unix -  -   -   -   8  smtp' in /etc/postfix/master.cf. If your system sees 16 CPUs in 'dmesg' command out, it's ok to increase 8 to 16. Note: the more '$max_servers', the more memory it needs. but 32GB is definitely ok.

4 (edited by zee 2015-02-15 18:19:49)

Re: Webaccess is down

The server shutting down was fixed after i reverted to the original nginx config.

Now I have another issue: after 8 hours of server uptime, the email server froze again and I had to reboot the machine. Any hints?

5

Re: Webaccess is down

Could you explain what you mean by it froze? Whole server/VPS froze, or just a process/program? [or functionality]

6

Re: Webaccess is down

By frozen i refer strictly to the mail server, not the entire VM.

IMAP, SMTP, POP, webclient weren't responding anymore until I rebooted the machine.

7

Re: Webaccess is down

Any chance you could let me [or Zhang] debug this via SSH? Its really strange to see them all go down at the same time, being all in all 4 different packages [nginx, php5-fpm, postfix, dovecot]

8

Re: Webaccess is down

Sure thing. Now the server has 6 hours uptime and everthing seems fine.
Contact me on Skype at tghiorghiu or via email: tudor.ghiorghiu@gmail.com

7t3chguy wrote:

Any chance you could let me [or Zhang] debug this via SSH? Its really strange to see them all go down at the same time, being all in all 4 different packages [nginx, php5-fpm, postfix, dovecot]

9

Re: Webaccess is down

I'm afraid this may be caused by Fail2ban. Someone had many password failures and this triggers Fail2ban to ban his/her IP address, which feels like "frozen".

Restart both iptables and fail2ban services should fix this issue. If you're experienced system admin, you can use iptables command to delete the blocked IP instead of removing all blocked IP addresses by restarting iptables+fail2ban services.

10

Re: Webaccess is down

It's not fail2ban, his nginx log gets spammed with php5fpm errors