1 - Make scripts_plugin email heuristic less stupid, or maybe even ask for an
2 email. This is tracked as Scripts #224 (this issue) and Scripts #193
3 (tracking a contact address).
4 - Current parallelization probably does a bad job distributing
5 working tasks over different components of the pipeline. Fix
6 this by adding jitter? Trying to smear things out?
8 - Test head doesn't do quite the right thing with version numbers
9 (shouldn't git describe, instead should give a version infinitely
11 - Strategy introspection and disabling.
12 - prepare-config (and others) create .wizard dir even
13 when not strictly necessary
14 - Bug out immediately if tags are not present in the master tip
16 - pending doesn't seem to get written out properly sometimes (or
17 it's being deleted); this makes it hard to --continue on the
18 event of an upgrade failure. Also, we seem to bounce back to
19 the production copy to check pending even when we run --continue
21 - Newline checks are /really really/ expensive on AFS; see if
22 we can minimize them or something. Right now, we're testing
23 a fix where we don't clone with --shared.
24 - Replace .split("\n") with .splitlines()
26 - Need to fix existing repo history? (not adding extra commits;
27 that'll be more difficult)
29 git rebase -i -p --root --onto COMMITID
31 This won't work if you need to change the very root of the
32 repository. You'll probably end up with conflicts and have
33 to manually resolve everything afterwards.
35 But usually you won't need --root --onto unless you really
36 fucked up the pristine branch. If you just need to change
39 git rebase -i -p COMMITID
43 - [SCRIPTS] MediaWiki 1.6.7, 1.9.3 and 1.10.0
46 "Connection to scripts.mit.edu closed" is confusing
47 the URL should be easier to copy and paste, which means we should
49 We should ... upgrade our autoinstaller
50 Apparently installing WordPress updates or themes never indicates
51 completion, and just says "Downloading..", and you have to guess
55 - phpBB or phpbb? (right now it's the former)
56 - need an upgrade story; srv needs more versions
57 - need a story about install/ contrib/
59 - Give users a "certificate" of their merge, which they can
60 use to reuse that merge commit if something unrelated fails.
62 - Human readable quota output
63 - Nice error message on --continue if you forgot to git add your
64 resolved file (look for conflict markers)
65 - The merge interface is a kind of major UI disaster; you won't
66 be able to use it unless you know how Git works. Also, the
67 merges can be quite difficult to resolve if upstream has made
68 large formatting changes like reindenting. We may also consider
69 providing a --rebase option, which seems to do better when
70 big problems like this show up.
72 - Better mass-* support for just one user (this includes letting
73 a user mass upgrade just his own autoinstalls)
75 - Show progress or something when upgrading
76 - Allow 'sticky notes' for future upgraders to notice
77 - .wizard/url semantics are subtly wrong: in particular, if we
78 explicitly configure a URL at install, we should be able to
79 detect this URL as baked in from the configuration
81 - Rerere support doesn't actually work
82 - "Version 3.0.0 doesn't exist; did you mean 3.0?"
83 - Be a little more intelligent when perform web checks; for example,
84 if we get a forbidden message, that probably means we go the right
85 address but it's blocked off; if we get a 404 message, that probably
86 means wrong address. Account Unknown is something particularly good
88 - Wordpress module can do something intelligent if we get redirected
89 to the installation page.
90 - wizardResolve* files seem to get left in tmp en-mass, and we don't
93 - [SCRIPTS] Wordpress needs to get rid of the siteurl hack, so that it
94 actually has a fully-qualified URL http://foo.scripts.mit.edu/blah.
95 This will also fix Wordpress's cron functionality. We should be
96 careful not to write over users who are on vhosts. We should figure
97 out who is still on twiddle paths. We should make sure the redirect
100 - Remerges aren't reflected in the parent files, so `git diff` output is
101 spurious. Not sure how to fix this w/o tree hackery.
102 - Sometimes users remove files. Well, if those files change, they automatically
103 get marked as conflicted. Maybe we should say for certain files "if they're
104 gone, they're gone forever"? What is the proper resolution?
106 - Parse output HTML for class="error" and give those errors back to the user (done),
107 then boot them back into configure so they can enter in something different
109 - [SCRIPTS] If you try to do an install on scripts w/o sql, it will sign
110 you up but fail to write the sql.cnf file. This sucks.
112 - [SCRIPTS] Web application for installing autoinstalls has a hard
113 problem with credentials (as well as installations that are not
114 conducted on an Athena machine.) We have some crazy ideas involving a
115 signed Java applet that uses jsch to SSH into athena.dialup and
119 - Tidy up common code in callAsUser and drop_priviledges in shell;
120 namely cooking up the sudo and environment variable lines
121 - Summary script should be more machine friendly, and should not
122 output summary charts when I increase specificity
123 - util.fetch() should use urllib under the hood, not httplib. Code
124 has to be changed. We should log if we get redirected.
127 - Add support for mypristine workflow
128 - Figure out why Sphinx sometimes fails to crossref :func: but wil
129 crossref :meth:, even though the dest is very clearly a function.
130 Example: :func:`wizard.app.php.re_var`
131 - The TODO extension for Sphinx doesn't properly force a full-rebuild
132 - Make single user mass-migrate work when not logged in as root. The
133 primary difficulty is making the parallel-find information easily
134 accessible to individual users: perhaps we can do a single-user
135 parallel-find on the fly.
136 - Don't use the scripts heuristics unless we're on scripts with the
137 AFS patch. Check with `fs sysname`
138 - Make 'wizard summary' generate nice pretty graphs of installs by date
139 (more histograms, will need to check actual .scripts-version files.)
140 - It should be able to handle installs like Django where there's a component
141 that gets installed in web_scripts and another directory that gets installed
144 - [SCRIPTS] ACLs is a starting point for sending mail to users, but it
145 has several failure modes:
146 - Old maintainers who don't care who are still on the ACL
147 - Private AFS groups that aren't mailing lists and that we can't get
148 to A question is whether or not sending mail actually helps us:
149 many users will probably have to come back to us for help; many
150 other users won't care.
151 - Whatever happens here should be used to improve user.email()
153 [ XXX: metadata.rst ]
155 COMMIT MESSAGE FIELDS:
157 Installed-by: username@hostname
158 Pre-commit-by: Real Name <username@mit.edu>
159 Upgraded-by: Real Name <username@mit.edu>
160 Migrated-by: Real Name <username@mit.edu>
161 Wizard-revision: abcdef1234567890
162 Wizard-args: /wizard/bin/wizard foo bar baz
166 Committer: Real Name <username@mit.edu>
167 Author: lockername locker <lockername@scripts.mit.edu>
171 [ XXX: doc/deps.rst ]
172 * Some parts of the infrastructure will not be touched, although I plan
173 on documenting them. Specifically, we will be keeping:
175 - get-homedirs.sh. which needs to be run as root on scripts. Store
176 in /mit/scripts/sec-tools/store/scriptslist
178 - parallel-find.pl, and the resulting
179 /mit/scripts/sec-tools/store/versions
181 [ XXX: doc/deploy.rst ]
182 GET APPROVAL BEFORE PROCEEDING ANY FURTHER;
183 THIS IS PUSHING THE CHANGES TO THE PUBLIC
185 NOTE: The following commands are to be run on not-backward.mit.edu.
186 You'll need to add daemon.scripts-security-upd to
187 scripts-security-upd to get bits to do this. Make sure you remove
188 these bits when you're done.
190 10. Run `wizard research appname`
191 which uses Git commands to check how many
192 working copies apply the change cleanly, and writes out a logfile
193 with the working copies that don't apply cleanly. It also tells
194 us about "corrupt" working copies, i.e. working copies that
195 have over a certain threshold of changes.
197 11. Run `wizard mass-upgrade appname`, which applies the update to all working
200 12. Run parallel-find.pl to update our inventory
202 [ XXX: doc/metadata.rst ]
203 * The repository for a given application will contain the following files:
205 - The actual application's files, as from the official tarball
207 - A .scripts directory, with the intent of holding Scripts specific files
208 if they become necessary.
210 - .scripts/dsn, overriding database source name