Ubuntu Servers :: Dovecot Binding To LDAP On Reboot?
Jun 9, 2011
Today I'm trying to configure Postfix+Dovecot to use Samba4's LDAP database for authorisation and mail delivery. As I can see from /var/log/mail.log, Dovecot tries to bind to LDAP right after reboot , but fails:
Code:
pdcadmin@PDC1:~$ cat /var/log/mail.log
Jun 9 13:06:46 PDC1 dovecot: auth(default): ldap_bind
Jun 9 13:06:46 PDC1 dovecot: auth(default): ldap_simple_bind
[code].....
Believing this to be a sign of succesfull bind, I couldn't understand the reason behind it. Why do I need to restart or reload dovecot service to make it work (though it fails on the next step with "dict_ldap_lookup: Search error 1: Operations error" and "451 4.3.0 ... Temporary lookup failure")?
So I'm doing some tests to see if I can create a LDAP server + Domain controler with Samba on Ubuntu 7.10. It might seem weird to use such an old release but I've seen a few guides on that particular version and I thought I'd give it a tr using this guide
url
So far I'm good until I modify the pam.d and nsswitch with auth-client, after reboot I'm unable to log in to the system and I'm forced to go into safe mode and restore the backups of those files.
What should I be looking for to prevent this from happening? I'm assuming that it's essential to modify these files in order to enable the LDAP authentication on the server.
I have Centos ( and Postfix+ldap+dovecot ) TLS works with Postfix and LDAP. When I open evolution mail client I can browse ldap tree and search for users, send-receive mails ...all fine
Few days ago we had a server maintenance. The system was shutdown, we fix the CPU fan, and start the system again. But somehow when the system starts, our IMAP server - dovecot is not running. It just sit like a rock. Because the machine had a CPanel/WHM, I tried to restart dovecot using cpanel and got a message:
Code:
That was not really useful....
When I tried to restart dovecot using command line, I got nothing. really nothing.
Code:
How to find out what happens to my IMAP/dovecot? And anyway to make it run again?
I have a server hosted with Hetzner in Germany and have had it there for over a year now without any glitches. Last week very abruptly they deactivated the server giving me a very vague reason of "I was using another IP that was not assigned to me". After several calls I had the server reactivated but am still puzzled by this and Hetzner Germany has been unable to provide me with any proof, logs, traces or anything for that matter to validate their claims.
My primary (physical) server has one IP allocated (static) and IPv6 is turned off. On top of this server I run 3 VMware servers, 1 Ubuntu, 1 Debian, 1 Windows and all 3 of these VM's have static IP's configured.
I've done traceroutes, netstats and looked through every single config file with a fine toothcomb without finding anything and the problem has not reoccured either.
What I would like to know is if there is some way that someone could craft an attack of some sort that could cause a server to do this, to latch onto another IP not assigned to its NIC?
Something that has been in the pipleline at work for a while is user-based web directories. Main PDCs are running Windows Server 2003 using Active Directory, ideally what would happen is that users have a web share under [URL].. - the server behind this would be Linux (either Fedora or CentOS).
What kind of configuration would be needed for Apache to make this possible? The way I have planned so far is to have the Linux box auth against the AD domain (possibly joined), with Apache setup to share local public_html folders. Not sure how I can get rid of the tilde from the start of the username, but it should be pretty easy.
School with linux running on students' laptops, connecting via wlan to a Debian NFS and LDAP server. Every student logs on his/her profile residing on the NFS server.The clients are set up with autofs. Earlier, I had set up the wireless network in /etc/network/interfaces, but this time I decided to configure network manager so as to bring up both wireless and wired network before logon. This setup has been working on for the last fire or five years with only minor changes. Also worked with Karmic Koala, but still with the interfaces file instead of networkmanager. The Vostro is also new here, we've previously used mostly Dell Latitude D505s.
So here is what works:
1: Clients can log on to LDAP and NFS servers both wired and wirelessly. Everything is smooth.
2: While on LAN, shutdown and restart works flawlessly (and quick as a breeze, I'm really impressed by startup/restart/shutdown times, under 25 secs!).
3: Shutdown and restart also works wirelessly when doing it either from a local account or from the GDM chooser.
What doesn't work, however, is shutting down or restarting directly from a networked account connected while only being connected over the wireless network. This is what's being displayed on the terminal after it has tried tho shut down for a while:
Code:
The system is going down for halt NOW!
acpid: exiting init: cron main process (1011) killed by TERM signal. init: tty1 main process (1365) killed by TERM signal.
[code]...
If I try ctrl-alt-del at this stage, it says:
"init: rc main process (3030) killed by TERM signal"
"Checking for running unattended-upgrades: "
And then it will hang again, until I hold the powerbutton for some seconds. The unattended-upgrades part is what seems to be the culprit. I suspect it is about the wireless network not being connected any longer or something like that, but I'm not sure about how to go about debugging shutdown scripts here. I'd be grateful for pointers. I will try and see how it goes with the old interfaces file setup, but I'd rather make nm work.
I am able to copy mails from another account to the account on my server. But I've got no clue of where dovecot stores them. Not in /var/mail I've checked..
This happened after I re-installed with a 9.10 Ubuntu Server disc.
I have a problem connecting my roundcube to my imap server.
So, here it is : I have a working imap server (dovecot), installed following the guide provided here [url]. I can successfully telnet it, or ssl it (from localhost or from another ip of my local network). I can even use my thunderbird from another local computer. Well, for me it is working fine.
Then I installed roundcube (0.3.1-3) with synaptic, but I cannot connect to my localhost imap server.
I have set auth as plain ($rcmail_config['imap_auth_type'] = 'plain';)
Looking in my dovecot log, i see two different behaviour. One is when it seems correct, and roundcube log in imap server, but is disconnect immediately :
Second is when server is refusing the password :
I am working with a plain text passwd, with no encryption at all. Does roundcube encrypt the password I type before to sent it to imap ?
I cannot explain the first behaviour, when roundcube sometimes managed to connect but disconnect immediately. I manage ONE time to reach my mailboxes, but i don't know how, and this never happens again.
I have tried to find solution in existing posts but could not specifically find any with my kind of issues and hence a new post on oft repeated subject !! -- and apologies for a long long post here.Here is where I am ..On a AMD 64bit machine - I have ubuntu 10.10 desktop installed. I want this development machine to support virtual mailboxes so that I can use them from multiple apps and create real life deployment situations.I installed postfix + dovecot following the tutorials available here and current state is - I can send mails using telnet sessions and I see that the mail files are getting created in /Maildir form as I have directed in the conf files. I have configured Thunderbird mail client as well.
Issue #1: Mail sending works from Thunderbird but it always responds back with 'No mail on server' message when I try to receive mails. SMTP is configured with STARTTLS and POP3 with None (i.e. plain text password)Issue #2: Also, while going thru conf, logs and during testing - I found a few things which defer in this installation for authentication. I have given the session transcripts here.Issue #3: That being major issue - I also want to configure my virtual users to use TB client to access their mails - I did not find any tutorials or pointers towards that in my search for past few days. If I send mails to a non-Unix virtual user - the mail gets stored into /home/vmail/<domain>/<user>/new directory.Here are the conf files.main.cf for postfix
Code: # See /usr/share/postfix/main.cf.dist for a commented, more complete version # Debian specific: Specifying a file name will cause the first
Dovecot is setup to run on each startup but each time I restart my system, I find that it is not running(killed), when I check my logs it says time moved backwards for x seconds. Essentially, I'm having this problem, [url] what is the solution in ubuntu so that dovecot would not be killed on startup? currently I have to put /etc/init.d/dovecot restart into rc.local to restart it again...
I have made some little mail server (postvix+dovecot+fetchmail) at ubuntu. The mail serve is using system accounts. Now I want to creat Shared mailbox. I have found some information, but I can't understand it correctly. Can U view the below information and correct me.
I have several accounts: Oleg (my account - /home/oleg/Maildir/); pankov (/home/pankov/Maildir); cv ( /home/cv/Maildir/). I want to share the mailbox of CV user: I have add the follwoing information to dovecot.conf:
# You need to create also a private namespace: namespace private { separator = / prefix =
I do not know why I could send email outside but not receive. I use relay to my ISP that port 25 is blocked by my ISP. What can I doing with configuration when it should be works to receive mail? Okay, I have sent a test e-mail between localhost and it works. I can send from mailserver to my gmail without problem. So why I cannot receive email from gmail?
I have a very basic install of dovecot and postfix on the latest version of Ubuntu server. This is an internal only email server with internal only DNS. I can send email via clients and check said emails via the command line when logged in as the appropriate user. I cannot for the life of me check said emails via SquirrelMail on the server nor using IMAP clients. I have no idea where to look and I can't find a basic tutorial for the life of me. Where should I start?
I have a system running at home that uses Getmail to retrieve mail from my ISP's pop server. Dovecot then offers that mail over IMAPs to my desktops running Thunderbird.
The reason I have resorted to using Getmail is because I don't have a static IP (from my ISP) for my server, and thus this server doesn't act as an MX.
I have implemented Spamassassin in my Getmail script as described here.
From further research, I understand that in order to fully utilize Spamassassin 's potential, I have to resort to training it with SA learn.
Currently I still receive spam messages, but 50% of spam is marked as ****SPAM****, and the other half is not marked at all.
My question is this:
1) How do I get getmail to move messages marked as spam by spamassasin to be moved to a JUNK folder within my mailbox automatically?
2) I thought of creating a folder where my users can move messages they deem to be spam, and set up a crontab script to invoke salearn regularly on this folder to get the bayes engine to learn from it. Is this the correct way of doing it?
recently I've replaced an old windows mailserver with a new one running 10.04 LTS (postfix - dovecot) but I got a strange problem with MS Outlook 2003 and 2007 on Win XP. These MUAs receive new messages extremely slow. Some of messages are received repeatedly. All I found in /var/log/mail.log is login and then (after 1 minute) disconnection for inactivity. No errors.:
Jul 11 10:36:51 dovecot: pop3-login: Login: user=(user@domain.org), method=NTLM, rip=xxx.xxx.xxx.xxx, lip=xxx.xxx.xxx.xxx Jul 11 10:37:07 dovecot: POP3(user@domain.org): Disconnected for inactivity top=0/0, retr=1/5434, del=0/12, size=954793 nothing more. And outlooks wait and wait reporting "receiving". What they are waiting for? Is there something dovecot should send them?
At the same time Mozilla Thunderbird and Opera Mail on the same network with the same setup don't have any problems. I just don't get it. Does anyone know what to do to get it work please?
Last time I rebooted, sendmail & dovecot started from init and everything worked. I needed to reboot this weekend [new videocard] & yum had given me a new kernel etc.Now.2.6.34.7-56.fc13.i686 running Thunderbird - with 2 accounts chuck@otherserver.com and chuck@localhost. The "otherserver" account works fine. Localhost won't pull in the mail.
We are planning on rolling out a mail service that will backend a webmail/groupware interface using postfix and dovecot, and we are trying to nail down an architecture for the system. I was hoping I could get some opinions from you good people on the different models we have come up with, and any improvements. To preface this discussion, we use NFS as the backend storage for mail so that clusters of mail servers can be created.
One possible solution is to separate three services, inbound email, outbound email, and message retrieval onto separate servers. We currently run a mail system with 65K users in this manner and it works, but there is a lot of wasted cpu on the inbound and outbound mail servers. where the dovecot servers are doing most of the work.
Another solution would be to combine the dovecot and inbound postfix servers to run dovecot as the local transport agent for postfix with LMTP, giving us quota and sieve capability.
The last solution is to run all three services on the same machine, giving us a single server image that we can clone on demand when we need to add more performance.
The single server, all three services, solution seems to be the simplest to build and manage, but putting all your eggs in one basket like that concerns us. What if there is a sudden influx of mail from our filtering platform, and it slows down all the other processes? What if the NFS server becomes disconnected from the network and the load average on each server skyrockets killing the ability for outbound mail to flow (where a separate outbound mail server would continue running despite this issue)?
I have an email server configured with postfix, dovecot, saslauthd and squirrelmail. It works great. In fact I'd even say it works too well: in squirrelmail, I can configure any sender address and my server will accept it. Message will be sent, wether the sender's address is [URL] or [URL]. Is there any way of limiting it in such way that only [URL] is accepted as sender's address?
I am new to to dovecot and would like some help, so please be kind.I have one user that can send mail but not read it either by pop3 or imap.I am running Dovecot 2.0.11 on Fedora 14.
I installed Postfix, Roundcube (webmail) and Dovecot. Postfix stores the e-mails in maildirs located at /home/postfix/maildir/[user]@[domain]. Now I set the mail_location configuration option in Dovecot to: mail_location = maildir:/home/postfix/maildir/%u@%d The problem I run into is logging in with [user]@[domain]. Right now I can only login with [user] (without @[domain]), which makes the mail_location invalid (no e-mails can be found). How could I configure dovecot to use the domain name as well as part of the username?
make Dovecot automatically archive the servers copy of the email if/when a client connects to it using POP? I am trying to achieve the the same thing as Gmail's feature.
I have recently set up a mail server running dovecot and everything is fine. However, I now want to retrieve the contents of 2 pop mail boxes and insert them into the IMAP structure under dovecot. I need to check for spam and sort them according to sender and recipient.
What is my best software option for retrieving the mail?
Is it even possible to use LDAP on Ubuntu 11.04? After a full day of googling, every guide I can find is either for another version of Ubuntu or is horribly broken (including the official docs).
