Temp mail

[bspitzbart]

Upon approval I will switch to @ipa alert lists (sot_ace_alert to @cfa (google) - sot_ace_alert@ipa does not exist) at 2 pm on 8/15 and backout to @Head (majordomo) at 9 am on 8/18. Alerts from OCC (c3po-v) can be sent to @ipa directly, bypassing the HEAD mail server during the power outage.

[bspitzbart]

Not sure why the old commits are showing up again or why the latest commits are put on June 6, but the diffs look correct.

[swolk]

Is this operational now?

On Fri, Aug 15, 2014 at 11:57 AM, bspitzbart notifications@github.com
wrote:

Not sure why the old commits are showing up again or why the latest
commits are put on June 6, but the diffs look correct.

—
Reply to this email directly or view it on GitHub
#3 (comment).

*-Scott *

---

swolk@cfa.harvard.edu

Chandra X-Ray Center
Harvard-Smithsonian
Center for Astrophysics | 617-496-7766
60 Garden St. MS-70 | 617-721-4360 (cell phone)
Cambridge, MA 02138 | 617-495-7356 (fax)
http://hea-www.harvard.edu/~swolk

[taldcroft]

@bspitzbart - I'm not sure what's happening with those commits, but you need to be a bit careful about merging this. Will you be around this evening to install? I could look into this but I need to be away from keyboard for another 45 mins.

[taldcroft]

It looks like #1 and #2 were never merged into master, or maybe it wasn't merged on github, or your local master has those PRs but it was never pushed to github? If you have those commits on your local master then do

git push origin master

and then these extra commits will probably go away in this PR.

[taldcroft]

As far as the content of 34a3a4f, this looks fine.

[bspitzbart]

I think the problem was that I did not resync the local repo after the last
pull to master.

git fetch origin # fetch all changes since no branch name provided
git checkout master
git merge origin/mastergit fetch origin # fetch all changes since no
branch name provided

So all the changes were there, but not in git 'notation'.

For installation, I just copied the affected files to the operational
area. Do we really want to do the pull to master since on Monday we
want the old (current) master?

I can come back to the office at around 6:30 if you have any concerns.

On Fri, Aug 15, 2014 at 4:49 PM, Tom Aldcroft notifications@github.com
wrote:

It looks like #1 #1 and #2
#2 were never merged into
master, or maybe it wasn't merged on github, or your local master has those
PRs but it was never pushed to github? If you have those commits on your
local master then do

git push origin master

and then these extra commits will probably go away in this PR.

—
Reply to this email directly or view it on GitHub
#3 (comment).

[jeanconn]

Brad, I'm not sure what "I think the problem was that I did not resync the local repo after the last
pull to master." means in this context, but if these changes

d25a60e
86d2cb3
bcba510

are already in the installed/running code, it would be good to get them into master on github.

I suppose one way to do that would be to copy your temp_mail branch, undo the last two changes, and then make that a new PR against master.... something like

git checkout temp_mail
# make a new branch with temp_mail as the start point
git checkout -b refix_ace_alerts 
# throw away the last two commits
git reset --hard bcba510       
# push up to a new feature branch on github
git push origin refix_ace_alerts 

and then make a new PR against master from refix_ace_alerts. If you go this way, you'll again want to rebase the temp_mail change against the new-plus-PR master after the PR is merged.

Or you could skip the pull request and just merge those 3 'missing' changes into master on github by doing something like

git checkout temp_mail
git checkout -b refix_ace_alerts
git reset --hard bcba510
git checkout master
# do a fast-forward merge of the 3 old changes into master
git merge --ff refix_ace_alerts
git push origin master

If you update master this way, you shouldn't need to rebase the temp_mail branch.

Tom, do you have ideas to make this cleaner? I don't think it would be cleaner to cherry-pick, but I'm not sure about nicer options than the reset in a throw-away refix_ace_alerts feature branch.

[taldcroft]

No need to push to master if it will be backed out.

At some point next week we should all (you, me, @jeanconn) go over workflow to get MTA exactly on track. Now that you have some real world experience things will make more sense.

[bspitzbart]

Thanks for all the help. I can meet sometime next week.

On Fri, Aug 15, 2014 at 5:57 PM, Tom Aldcroft notifications@github.com
wrote:

No need to push to master if it will be backed out.

At some point next week we should all (you, me, @jeanconn
https://github.com/jeanconn) go over workflow to get MTA exactly on
track. Now that you have some real world experience things will make more
sense.

—
Reply to this email directly or view it on GitHub
#3 (comment).