Full IRC transcript here
Jared (1st day back from paternity leave)
- Josie and I are both back to work, me starting more officially as of
today. The kid is still here, no day care, so current plan is for
slightly extended days to deal with going briefly offline during
crying jags, whatnot.
- I'm seriously considering just doing part time, 3 days per week, as
Josie is going into her office 2 days a week, and it's unrealistic to
think I'm getting anything done during those days.
- Work priority is to establish and work through server logistics,
specifically shutting down our SF colo at 365 main and 543 howard.
- 543 howard of course means tinderbox and build issues, as i've
flagged. the big box is at 365 main, as is bugzilla, wiki, web sites,
downloads, lab, backups, archives/dashboard.
- Basic plan is to move all 365 main colo to our donated colo at isc/
redwood city, shutting down a number of current isc machines.
<<Jeffrey and Grant offer to help with move>>
thanks, the hands might help.
- This will probably happen mid-day, just to coincide with donated-
colo's workday schedule.
- We will need to move svn to a 365 colo box probably, then move that
box to redwood city.
- Will have a punchlist of stuff, and work through it until i'm out of
work, i suppose. (It won't be maintained in chandler probably, as i
need a highly-ordered work sequence).
- Will take back all server maintenance from folks, as desired.
listmaster/blogmaster too.
- The two mail servers are ones that might need to move or get shut
down to make space for the from-365-main servers. It would be best,
long-term, if we shut down osaf email, or had it hosted somewhere
else. that's a little daunting.
<<JeffreyH>> could we at least keep the lists under our control?
- It would be hard to have local mailing lists if we don't have a
local mx, etc.
the basic issue is we currently have 8-9 U of rack space in use by
ISC. We're their biggest "customer". At 365 main, I have a 4U Hub box
and 2x 2U servers. So first approximation, every ISC box should get
shut down and services migrated elsewhere. I can try to dig up VM
space on those to move mail to, but I worry about long-term
maintenance. Hosting mail for friends used to be fun, but certainly no
longer; it's a real job.
- Most of the hassle is spam management, so if someone else MXes, and
we just accept email from them, running mailman isn't too much of a
hassle, if we get listmaster volunteers.
- Will probably recommend more aggressive "mail just bounces if you're
not subscribed" policies, coupled with some volunteer for chandler-
users if we want to leave it open.
<<Agreement from gbaillie, JeffreyH>>
There was also some discussion of shutdown of build servers at 543
Howard, see the transcript for more.
Randy:
-testing fix for weird tz issue (fingers crossed), watching hub today
to see if issue pops up again
-more testing, focusing on caldav and other clients today
-continuing to add/update docs
Grant:
- Committed recurrence dialog changes, and fired off and announced the
test build.
- Fixed Bug 12239 (Drag-and-drop to minicalendar doesn't work on Linux).
- Fixed an issue Rick Rawson noticed in test build ("Custom..." not
working if recurrence is "Once")
- Started going over 0.7.future bug list (mostly just punting).
- Implemented "1/2/3/4/5/6/7 days from the end of the month" UI at
Rick's suggestion; noticed vobject bug 12245 (Doesn't serialize
_bynmonthday in rrules)
- Trying to reproduce Bug 11266 (Chandler won't launch because
repository says it's in use by another Chandler after rebooting) by
doing evil things to Chandler running on my windows laptop; no success
so far.
Jeffrey:
- I did a bunch of vobject work yesterday, andwas about to spin a
release
when Grant pointed out a recurrence bug (good catch!)
- just writing in test coverage for it, easy enough to fix
- then I'll spin a vobject release
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev
