[Piuparts-commits] rev 916 - trunk
Holger Levsen
holger at alioth.debian.org
Sun Jul 10 10:23:34 UTC 2011
Author: holger
Date: 2011-07-10 10:23:33 +0000 (Sun, 10 Jul 2011)
New Revision: 916
Modified:
trunk/TODO
Log:
update
Modified: trunk/TODO
===================================================================
--- trunk/TODO 2011-07-10 10:13:16 UTC (rev 915)
+++ trunk/TODO 2011-07-10 10:23:33 UTC (rev 916)
@@ -27,13 +27,6 @@
the bts and remove the "bugged" directory-status tracking
altogether. also remove "fixed".
-- with that it's possible to just delete any logs older than
- $time (4/8/12 weeks?) periodically. Probably worth a new config
- option. Another approch would be for each logfile to check if
- the package still exist in $distro and if not, delete it. But
- that uses way more CPU+IO, also it's (almost always) useful to
- retest packages anyway.
-
- use libeatmydata - either install it in all chroots or find a
better way to use it.
@@ -59,9 +52,11 @@
binary package. when doing so keep in mind the two tango icons
and credit those in debian/copyright.
-- publish FAI classes to setup piuparts.$fqdn automatically
+- once that is done, move to git.
+- publish FAI classes to setup piuparts.$fqdn automatically.
+
for 0.44:
- PackagesDB: get rid of 50% of unique()
@@ -83,12 +78,6 @@
for 0.45:
-- untestable packages are not recognicable as such, see
- http://piuparts.debian.org/source/a/apache2.html atm. (log is
- N/A, even though its there, even if empty) -
- http://piuparts.debian.org/sid/state-cannot-be-tested.html aint
- helpful neither.
-
- slave_run: cleanup stale proc mountpoints
- the templates used by update-reports.py should be taken from
@@ -178,7 +167,7 @@
- Kill children after a timeout to make sure the test doesn't run
forever.
-- piuparts-master: keep track of whom a reservation is made
+- piuparts-master: keep track of to whom a reservation was given
- piuparts-slave: timestamps to log messages
More information about the Piuparts-commits
mailing list