Re: Sync failing. Possible db corruption

2020-09-16 Thread Michael Sofka
, but they appear to be from turning on verbose logging, and the synchronization being out of step from the previous errors.  As people check mailboxes, it is catching up. Mike On 9/15/20 4:20 PM, Michael Sofka wrote: I have confirmed that running sync_client on individual mailboxes works, as does

Re: Sync failing. Possible db corruption

2020-09-15 Thread Michael Sofka
I have confirmed that running sync_client on individual mailboxes works, as does running sync on the accumulated log files.  But rolling replication is not working.   After sync_client is restarted there is a burst of sync activity on the accumulated log, then nothing.  The new log file

Re: Sync failing. Possible db corruption

2020-09-15 Thread Michael Sofka
Okay, these errors may be admin error. In running the old logs I specified -u, instead of -m. I am still concerned about the DBERROR, but those did not appear after I restarted cyrus.  And rolling replication is still not working. Mike On 9/15/20 10:49 AM, Michael Sofka wrote: Sep 15 10

Sync failing. Possible db corruption

2020-09-15 Thread Michael Sofka
Before I proceed, I am seeking advice. This is on Cyrus IMAP Murder v2.4.18-Debian-2.4.18-3 On one of our back-end servers sync started failing.  I restarted sync_server on the replication, and on the backend server, which did not work.  There are the following errors in the logs Sep 15

Re: Thunderbird and "Login to server failed".

2017-10-05 Thread Michael Sofka
This was one of the symptoms of slow authentication that we saw before I increased the saslauthd threads. Mike On 10/04/2017 09:09 AM, Eugene M. Zheganin wrote: Hi, I'm using cyrus-imapd (now 2.4.x) and Thunderbird. The problem is, when I click fast on the e-mails, I often get the "Login to

Re: Using user_deny.db

2017-09-19 Thread Michael Sofka
On 09/19/2017 10:28 AM, Ken Murchison wrote: I believe that is it prior to authentication, based on my notes: https://lists.andrew.cmu.edu/pipermail/info-cyrus/2010-June/033119.html user_deny.db is NOT checked prior to completion of LOGIN authentication, although it probably could/should. 

Re: Bad logins bogging down server

2017-09-19 Thread Michael Sofka
On 09/19/2017 10:12 AM, Dan White wrote: The botnet is still hammering away, checking those old accounts.  But the bottleneck appears to have been saslauthd threads.  Doubling the thread count from 5 to 10 has resolved the problem for now.  (And, If you're comfortable with caching, increase

Using user_deny.db

2017-09-19 Thread Michael Sofka
We have many recalcitrant, bad, accounts constantly checking IMAP, long after the student has graduated. I would like to use user_deny.db to simply tell them to go away. First, would this offer an advantage? That is, does "login" check user_deny.db before authenticating, or after? Second,

Re: Bad logins bogging down server

2017-09-19 Thread Michael Sofka
Follow up The botnet is still hammering away, checking those old accounts. But the bottleneck appears to have been saslauthd threads. Doubling the thread count from 5 to 10 has resolved the problem for now. (And, might even explain the occasional slow response from IMAP I've observed.)

Re: What is the state of Master-Master replication?

2017-07-28 Thread Michael Sofka
On 07/26/2017 04:54 PM, Michael Sofka wrote: A while back there was some discussion of supporting Master-Master replication in Cyrus. I'm busy updating from 2.4.17 to 3.0.2. What is the state of Master-Master, as opposed to Master-Replica replications? My current configuration is a Murder

What is the state of Master-Master replication?

2017-07-26 Thread Michael Sofka
A while back there was some discussion of supporting Master-Master replication in Cyrus. I'm busy updating from 2.4.17 to 3.0.2. What is the state of Master-Master, as opposed to Master-Replica replications? My current configuration is a Murder cluster with three front-end servers, two

Re: Cyrus aggregate compatibility.

2015-04-20 Thread Michael Sofka
On 2015-04-20 17:16, k...@rice.edu wrote: On Mon, Apr 20, 2015 at 05:11:00PM -0400, Michael D. Sofka wrote: Under the scenario, would 2.5 work better? Mike Hi Mike, In our case, the unconstrained I/O caused by the mandatory mailbox format conversion on first use would have necessitated

Re: Updating front-end and master server

2010-11-25 Thread Michael Sofka
Yes, that was a typo. All back ends are 2.3.16. -- Michael D. Sofka sof...@rpi.edu CMT Sr. Systems Programmer, Email, HPC, TeX, Epistemology Rensselaer Polytechnic Institute, Troy, NY. http://www.rpi.edu/~sofkam/ ==Original message text=== On Wed, 24

Multiple partition question.

2007-09-28 Thread Michael Sofka
Hello, We have a failing back-end server. The cause is, possible, the RAID controller. So, I'm working on migrating mailboxes off of this server, and on to another back-end server. The other backend server, however, does not have enough space. Since the target back-end server has SAN attached

Re: Mailbox subscriptions in a murder

2007-04-10 Thread Michael Sofka
=== On Fri, 06 Apr 2007 16:52:07 EDT Eli Ben-Shoshan wrote: Michael Sofka wrote: We do this with cyrus 2.2.12 (2 frontend, 1 master, 3 backend server) no problem. Some of the shared mailboxes are on one server, while subscribers are on another. We run the frontend servers in proxy mode only. Can

Re: shared mailboxes with virtual domain setup

2007-04-06 Thread Michael Sofka
We use shared mailboxes with Outlook and Thunderbird, and many other clients. Outlook and T-Bird don't have ACL editors, but Cyrus returns the list of shared mailboxes a user has access to. We recomend our Webmail client for ACL editing. We don't use virtual domains, so I can't say if that

Re: Mailbox subscriptions in a murder

2007-04-06 Thread Michael Sofka
We do this with cyrus 2.2.12 (2 frontend, 1 master, 3 backend server) no problem. Some of the shared mailboxes are on one server, while subscribers are on another. We run the frontend servers in proxy mode only. Mike -- Michael D. Sofka Sr. Systems Programmer, Postmaster pro tem Rensselaer