Context Navigation

Timeline

10:00 PM Changeset [1301] by geofft: Actually unblock reuter from sending (not just receiving) mail [help.mit.edu #985700]
2:35 PM Changeset [1300] by geofft: Unblock golem from receiving mail (discussed with him over zephyr)
2:32 PM Changeset [1299] by geofft: Unblock reuter from sending mail [help.mit.edu #985700]
5:06 AM Changeset [1298] by mitchb: Deploy Ruby on Rails autoinstaller (Merge of r1294:1297 from branches/locker-dev to trunk)
4:52 AM Changeset [1297] by mitchb: Final pre-deploy touchups to the Ruby on Rails autoinstaller o Don't use a custom ErrorDocument that implies all problems are Rails. o Tell users what directory their app is in. o Tell users to ssh to scripts to run the Rails tools. o Edit the initial welcome page to avoid talking about database creation and sqlite.

2:34 PM Changeset [1296] by mitchb: fedora-ds is now 389-ds - "People like alphanumeric digits"

5:44 AM Changeset [1295] by mitchb: Ruby on Rails autoinstaller! (but I...)

8:54 PM Changeset [1294] by mitchb: We got nss-ldapd into Fedora! Stop building it from OpenSuSE.
6:50 AM Changeset [1293] by mitchb: Certificate and Apache config for metu.mit.edu

10:00 AM Changeset [1292] by mitchb: Weight r-m the same as p-b and b-b because they're all uniprocessors LVS shows that r-m is consistently handling far more connections than p-b or b-b, as it should since we've been weighting them by amount of RAM. Unfortunately, the AMD SMP hangs have caused us to make these VMs uniprocessors, and I suspect that r-m's repeated brief SVN hiccups are due to handling more load than it's prepared for. So, until we can run SMPs again, number of virtual cores trumps amount of RAM.

10:54 AM Changeset [1291] by mitchb: Stop using the Fedora mirrors; they suck The mirrors lag behind the package release announcements by a disappointing time window, and MirrorManager is notorious for picking bad/slow/out-of-date mirrors. We'll use one of Fedora's masters for now.
10:51 AM Changeset [1290] by mitchb: Put -updates and -updates-testing repo configs under version control

11:21 PM Ticket #96 (parallel-find.pl will fail to find new-style autoinstalls) closed by ezyang: fixed: Fixed.
1:04 AM Changeset [1289] by mitchb: OpenSUSE bumped the nss-ldapd release again? Shocking! 6.4->6.5

9:29 PM Changeset [1288] by mitchb: Revert r1287 - we're not convinced the change is beneficial
8:46 AM Changeset [1287] by quentin: Don't execute scripts outside of /afs, unless they're in the trusted directory

6:16 AM Changeset [1286] by mitchb: AFS restarts are finished by 6:30; don't complain about load during them

3:20 AM Changeset [1285] by ezyang: Implement searching for Wizard autoinstalls.

7:16 AM Ticket #104 (Improve for-each-server script) created by mitchb: See RT ticket …
5:08 AM Changeset [1284] by mitchb: Do not let parallel-find skip non-athena-cell lockers We have many users in the sipb cell; our responsibility to them is no different. We need to scan all cells we have credentials for.

1:01 AM Changeset [1283] by mitchb: Stop scriptsifying rubygem-rails; Fedora has taken our patch
12:58 AM Changeset [1282] by mitchb: Update nss-ldapd from version 0.6.10 release 6.2 to release 6.4 OpenSUSE keeps changing the version every time I want to build a package

2:09 AM Changeset [1281] by mitchb: Fix spacing and detabify django autoinstaller I tried to convince myself not to care all day and failed; don't mind me.
1:29 AM Changeset [1280] by geofft: Little script to ssh to each active LVS realserver in turn

1:52 AM Changeset [1279] by mitchb: Stop telling Django users to touch index.fcgi; it never worked anyway
1:34 AM Changeset [1278] by mitchb: Make Django autoreloads actually work, without even touching index.fcgi

11:11 AM Changeset [1277] by ezyang: Add missing build dependency, as suggested by adehnert.

4:43 AM Changeset [1275] by mitchb: Temporarily scriptsify rubygem-rails to work around Fedora packaging error Between the version of this package appearing in Fedora 9 and Fedora 11, the spec file gained a stanza that removes all zero-length files from the destination tree. This actually removes four required zero-length files (and nothing else), without which running 'rails' to create a new application errors out. See https://bugzilla.redhat.com/show_bug.cgi?id=496480

Note: See TracTimeline for information about the timeline view.