Year: 2005
-
This weeks update
—
by
Well, we got a lot of things done this week since it was so quiet. At times we were a little too brave and broke a few things when we should not have, but in the end a lot was accomplished. Reducing duplicate email messages – We put in place a procmail script that checks…
-
Postifx and Authentication
—
by
After updating one of our 2 postfix mailservers we discovered that some email clients can’t deal with all the authentication options our mailserver presents to them. For example Eudora has an issue if postfix says it can do CRAM-MD5 as in the following example: 250-AUTH PLAIN DIGEST-MD5 LOGIN CRAM-MD5 250-AUTH=PLAIN DIGEST-MD5 LOGIN CRAM-MD5 For some…
-
Heartbeat on Fedora Core 4
—
by
Here are the steps I took to get heartbeat ( from linux-ha ) to run on Fedora Core 4 (fc4). It is by no means a clean install, but it appears to work for us at the moment. First download the RPM’s from http://www.ultramonkey.org/download/heartbeat/. I am using version 2.0.2 today. Then I installed curl to…
-
Friday Maintenance
—
by
For the past few weeks the load on our email servers, specifically the ones responsible for imap/pop/webmail, have been unable to keep up with the increasing demand. Thus causing access to imap/pop/webmail to be unavailable during times of peek usage. Delivery of email was not affected. To help alleviate the issue, we have spent the…
-
Friday’s Upgrades
—
by
Every day it seems like we are sending and recieving more and more email. That means every day our email system is more and more loaded. A couple of weeks ago we hit the limit as to how much email our disks could handle for everyone’s inboxes. We store all the users inboxes on a…
-
Horde, IMP, and imapproxy
—
by
So it turns out that the cache in imapproxy that we turned on yesterday has a problem. First of all, it seemed to work great. I think webmail was more responsive and there was a reduced load on our mail disks. However not all was well. It seems like some times when you deleted a…
-
Email performance
—
by
As many people have noticed, email performance during our peak usage time (9:50am-10:20am) has been extremely poor. The main reason we believe is that the disk drives our email inboxes are stored on are just not fast enough. Currently we are using a Sun T3 disk array with a 1Gig cache and 9 fibre-channel disk…
-
Replaced Switch
—
by
One of our HP 4000 switches had a bad fan that needed to be replaced before the entire thing overheated and died. Here is what Joel had to say about it: Dan and I replaced a switch in the Jackson Campus Center this morning. This switch is used by the Food Service, Health Service, Book…
-
Automatic Proxy Settings
—
by
We just figured out that somewhere along the line, we broke automatic proxy configuration stuff. To fix it there are a couple of things we can do. First we added this to our /etc/dhcpd.conf file: option WPAD code 252 = string; option WPAD “http://www.gac.edu/cgi-bin/proxies”; In our old config we had option option-252 “http://www.gac.edu/cgi-bin/proxies”; but that…
-
Strange Network Traffic
—
by
For some reason, since around midnight last night, there was a ton of broadcast traffic on our res-hall network. Around 150KB/s in fact. It was enough that wireless in my office was mostly unusable. Using tethereal I could see almost all of the traffic was coming from 2 MAC addresses and it was broadcast ARP…