1 - Test head doesn't do quite the right thing with version numbers
2 (shouldn't git describe, instead should give a version infinitely
4 - Strategy introspection and disabling.
5 - prepare-config (and others) create .wizard dir even
6 when not strictly necessary
7 - Bug out immediately if tags are not present in the master tip
9 - pending doesn't seem to get written out properly sometimes (or
10 it's being deleted); this makes it hard to --continue on the
11 event of an upgrade failure. Also, we seem to bounce back to
12 the production copy to check pending even when we run --continue
14 - Newline checks are /really really/ expensive on AFS; see if
15 we can minimize them or something. Right now, we're testing
16 a fix where we don't clone with --shared.
17 - Replace .split("\n") with .splitlines()
19 - Need to fix existing repo history? (not adding extra commits;
20 that'll be more difficult)
22 git rebase -i -p --root --onto COMMITID
24 This won't work if you need to change the very root of the
25 repository. You'll probably end up with conflicts and have
26 to manually resolve everything afterwards.
28 But usually you won't need --root --onto unless you really
29 fucked up the pristine branch. If you just need to change
32 git rebase -i -p COMMITID
36 - [SCRIPTS] MediaWiki 1.6.7, 1.9.3 and 1.10.0
39 "Connection to scripts.mit.edu closed" is confusing
40 the URL should be easier to copy and paste, which means we should
42 We should ... upgrade our autoinstaller
43 Apparently installing WordPress updates or themes never indicates
44 completion, and just says "Downloading..", and you have to guess
48 - phpBB or phpbb? (right now it's the former)
49 - need an upgrade story; srv needs more versions
50 - need a story about install/ contrib/
52 - Give users a "certificate" of their merge, which they can
53 use to reuse that merge commit if something unrelated fails.
55 - Human readable quota output
56 - Nice error message on --continue if you forgot to git add your
57 resolved file (look for conflict markers)
58 - The merge interface is a kind of major UI disaster; you won't
59 be able to use it unless you know how Git works. Also, the
60 merges can be quite difficult to resolve if upstream has made
61 large formatting changes like reindenting. We may also consider
62 providing a --rebase option, which seems to do better when
63 big problems like this show up.
65 - Better mass-* support for just one user (this includes letting
66 a user mass upgrade just his own autoinstalls)
68 - Show progress or something when upgrading
69 - Allow 'sticky notes' for future upgraders to notice
70 - .wizard/url semantics are subtly wrong: in particular, if we
71 explicitly configure a URL at install, we should be able to
72 detect this URL as baked in from the configuration
74 - Rerere support doesn't actually work
75 - "Version 3.0.0 doesn't exist; did you mean 3.0?"
76 - Be a little more intelligent when perform web checks; for example,
77 if we get a forbidden message, that probably means we go the right
78 address but it's blocked off; if we get a 404 message, that probably
79 means wrong address. Account Unknown is something particularly good
81 - Wordpress module can do something intelligent if we get redirected
82 to the installation page.
83 - wizardResolve* files seem to get left in tmp en-mass, and we don't
86 - [SCRIPTS] Wordpress needs to get rid of the siteurl hack, so that it
87 actually has a fully-qualified URL http://foo.scripts.mit.edu/blah.
88 This will also fix Wordpress's cron functionality. We should be
89 careful not to write over users who are on vhosts. We should figure
90 out who is still on twiddle paths. We should make sure the redirect
93 - Remerges aren't reflected in the parent files, so `git diff` output is
94 spurious. Not sure how to fix this w/o tree hackery.
95 - Sometimes users remove files. Well, if those files change, they automatically
96 get marked as conflicted. Maybe we should say for certain files "if they're
97 gone, they're gone forever"? What is the proper resolution?
99 - Parse output HTML for class="error" and give those errors back to the user (done),
100 then boot them back into configure so they can enter in something different
102 - [SCRIPTS] If you try to do an install on scripts w/o sql, it will sign
103 you up but fail to write the sql.cnf file. This sucks.
105 - [SCRIPTS] Web application for installing autoinstalls has a hard
106 problem with credentials (as well as installations that are not
107 conducted on an Athena machine.) We have some crazy ideas involving a
108 signed Java applet that uses jsch to SSH into athena.dialup and
112 - Tidy up common code in callAsUser and drop_priviledges in shell;
113 namely cooking up the sudo and environment variable lines
114 - Summary script should be more machine friendly, and should not
115 output summary charts when I increase specificity
116 - util.fetch() should use urllib under the hood, not httplib. Code
117 has to be changed. We should log if we get redirected.
120 - Add support for mypristine workflow
121 - Figure out why Sphinx sometimes fails to crossref :func: but wil
122 crossref :meth:, even though the dest is very clearly a function.
123 Example: :func:`wizard.app.php.re_var`
124 - The TODO extension for Sphinx doesn't properly force a full-rebuild
125 - Make single user mass-migrate work when not logged in as root. The
126 primary difficulty is making the parallel-find information easily
127 accessible to individual users: perhaps we can do a single-user
128 parallel-find on the fly.
129 - Don't use the scripts heuristics unless we're on scripts with the
130 AFS patch. Check with `fs sysname`
131 - Make 'wizard summary' generate nice pretty graphs of installs by date
132 (more histograms, will need to check actual .scripts-version files.)
133 - It should be able to handle installs like Django where there's a component
134 that gets installed in web_scripts and another directory that gets installed
137 - [SCRIPTS] ACLs is a starting point for sending mail to users, but it
138 has several failure modes:
139 - Old maintainers who don't care who are still on the ACL
140 - Private AFS groups that aren't mailing lists and that we can't get
141 to A question is whether or not sending mail actually helps us:
142 many users will probably have to come back to us for help; many
143 other users won't care.
144 - Whatever happens here should be used to improve user.email()
146 [ XXX: metadata.rst ]
148 COMMIT MESSAGE FIELDS:
150 Installed-by: username@hostname
151 Pre-commit-by: Real Name <username@mit.edu>
152 Upgraded-by: Real Name <username@mit.edu>
153 Migrated-by: Real Name <username@mit.edu>
154 Wizard-revision: abcdef1234567890
155 Wizard-args: /wizard/bin/wizard foo bar baz
159 Committer: Real Name <username@mit.edu>
160 Author: lockername locker <lockername@scripts.mit.edu>
164 [ XXX: doc/deps.rst ]
165 * Some parts of the infrastructure will not be touched, although I plan
166 on documenting them. Specifically, we will be keeping:
168 - get-homedirs.sh. which needs to be run as root on scripts. Store
169 in /mit/scripts/sec-tools/store/scriptslist
171 - parallel-find.pl, and the resulting
172 /mit/scripts/sec-tools/store/versions
174 [ XXX: doc/deploy.rst ]
175 GET APPROVAL BEFORE PROCEEDING ANY FURTHER;
176 THIS IS PUSHING THE CHANGES TO THE PUBLIC
178 NOTE: The following commands are to be run on not-backward.mit.edu.
179 You'll need to add daemon.scripts-security-upd to
180 scripts-security-upd to get bits to do this. Make sure you remove
181 these bits when you're done.
183 10. Run `wizard research appname`
184 which uses Git commands to check how many
185 working copies apply the change cleanly, and writes out a logfile
186 with the working copies that don't apply cleanly. It also tells
187 us about "corrupt" working copies, i.e. working copies that
188 have over a certain threshold of changes.
190 11. Run `wizard mass-upgrade appname`, which applies the update to all working
193 12. Run parallel-find.pl to update our inventory
195 [ XXX: doc/metadata.rst ]
196 * The repository for a given application will contain the following files:
198 - The actual application's files, as from the official tarball
200 - A .scripts directory, with the intent of holding Scripts specific files
201 if they become necessary.
203 - .scripts/dsn, overriding database source name