5 - The calling web code invocations are a mess, with stubs living
6 in the install, deploy modules and the real deal living in util. Furthermore,
7 we use the scripts-specific heuristic to determine where the app
8 lives, and the only reason my test scripts work is because they
9 get manually fed the domain and path by my environment variables.
11 Use system similar to database, with option for explicit override,
12 but otherwise attempting to determine from ambient code
14 - If you try to do an install on scripts w/o sql, it will sign you up but fail to write
15 the sql.cnf file. This sucks.
17 - wizard install wordpress should ask for password. One problem with this is that
18 Wordpress will still send mail with the wrong username and password, so Wordpress
19 will need to be patched to not do that. Alternatively we can initally set the admin
20 email to a null address and then fix it manually.
21 - --raw parameter for install which means an arbitrary commit can be installed
22 - Parse output HTML for class="error" and give those errors back to the user (done),
23 then boot them back into configure so they can enter in something different
24 - Get rid of our custom sizing code and use dialog's built-in sizing (i.e. width=0, height=0).
25 Maybe our sizing code is superior, maybe not.
27 - Replace gaierror with a more descriptive name (this is a DNS error)
29 - Pre-emptively check if daemon/scripts-security-upd
30 is not on scripts-security-upd list (/mit/moira/bin/blanche)
32 - Web application for installing autoinstalls has a hard problem
33 with credentials (as well as installations that are not conducted
34 on an Athena machine.) We have some crazy ideas involving a signed
35 Java applet that uses jsch to SSH into athena.dialup and perform
39 - Genericize callAsUser and drop_priviledges in shell
40 - Summary script should be more machine friendly, and should not
41 output summary charts when I increase specificity
42 - Summary script should do something intelligent when distinguishing
43 between old-style and new-style installs
44 - Report code in wizard/command/__init__.py is ugly as sin. Also,
45 the Report object should operate at a higher level of abstraction
46 so we don't have to manually increment fails. (in fact, that should
47 probably be called something different). The by-percent errors should
49 - Move resolutions in mediawiki.py to a text file? (the parsing overhead
51 - If a process is C-ced, it can result in a upgrade that has
52 an updated filesystem but not updated database. Make this more
54 - PHP end of file allows omitted semicolon, can result in parse error
55 if merge resolutions aren't careful. `php -l` can be a quick stopgap
58 - Figure out why Sphinx sometimes fails to crossref :func: but wil
59 crossref :meth:, even though the dest is very clearly a function.
60 Example: :func:`wizard.app.php.re_var`
61 - The TODO extension for Sphinx doesn't properly force a full-rebuild
63 - Make single user mass-migrate work when not logged in as root. The
64 primary difficulty is making the parallel-find information easily
65 accessible to individual users: perhaps we can do a single-user
66 parallel-find on the fly.
67 - Don't use the scripts heuristics unless we're on scripts with the
68 AFS patch. Check with `fs sysname`
69 - Make 'wizard summary' generate nice pretty graphs of installs by date
70 (more histograms, will need to check actual .scripts-version files.)
71 - It should be able to handle installs like Django where there's a component
72 that gets installed in web_scripts and another directory that gets installed
74 - ACLs is a starting point for sending mail to users, but it has
75 several failure modes:
76 - Old maintainers who don't care who are still on the ACL
77 - Private AFS groups that aren't mailing lists and that we
79 A question is whether or not sending mail actually helps us:
80 many users will probably have to come back to us for help; many
81 other users won't care.
83 PULLING OUT CONFIGURATION FILES IN AN AUTOMATED MANNER
85 advancedpoll: Template file to fill out
86 django: Noodles of template files
87 gallery2: Multistage install process
89 mediawiki: One-step install process
90 phpbb: Multistage install process
91 phpical: Template file
94 wordpress: Multistage install process
96 COMMIT MESSAGE FIELDS:
98 Installed-by: username@hostname
99 Pre-commit-by: Real Name <username@mit.edu>
100 Upgraded-by: Real Name <username@mit.edu>
101 Migrated-by: Real Name <username@mit.edu>
102 Wizard-revision: abcdef1234567890
103 Wizard-args: /wizard/bin/wizard foo bar baz
107 Committer: Real Name <username@mit.edu>
108 Author: lockername locker <lockername@scripts.mit.edu>
112 - It is not required nor expected for update scripts to exist for all
113 intervening versions that were present pre-migration; only for it
114 to work on the most recent migration.
116 - Currently all repositories are initialized with --shared, which
117 means they have basically ~no space footprint. However, it
118 also means that /mit/scripts/wizard/srv MUST NOT lose revs after
123 * Some parts of the infrastructure will not be touched, although I plan
124 on documenting them. Specifically, we will be keeping:
126 - parallel-find.pl, and the resulting
127 /mit/scripts/.htaccess/scripts/sec-tools/store/scriptslist
129 * The new procedure for generating an update is as follows:
130 (check out the mass-migration instructions for something in this spirit,
131 although uglier in some ways; A indicates the step /should/ be automated)
133 0. ssh into not-backward, temporarily give the daemon.scripts-security-upd
134 bits by blanching it on system:scripts-security-upd, and run parallel-find.pl
136 1. [ see doc/upgrade.rst ]
138 [ENTER HERE FROM CREATING A NEW REPO]
140 9. Push all of your changes in a public place, and encourage others
141 to test, using --srv-path and a full path.
143 [ XXX: doc/deploy.rst ]
144 GET APPROVAL BEFORE PROCEEDING ANY FURTHER;
145 THIS IS PUSHING THE CHANGES TO THE PUBLIC
147 NOTE: The following commands are to be run on not-backward.mit.edu.
148 You'll need to add daemon.scripts-security-upd to
149 scripts-security-upd to get bits to do this. Make sure you remove
150 these bits when you're done.
152 10. Run `wizard research appname`
153 which uses Git commands to check how many
154 working copies apply the change cleanly, and writes out a logfile
155 with the working copies that don't apply cleanly. It also tells
156 us about "corrupt" working copies, i.e. working copies that
157 have over a certain threshold of changes.
159 11. Run `wizard mass-upgrade appname`, which applies the update to all working
162 12. Run parallel-find.pl to update our inventory
164 [ XXX: doc/upgrade.rst ]
165 * For mass importing into the repository, there are a few extra things:
167 * When mass producing updates, if the patch has changed you will have to
168 do a special procedure for your merge:
170 git checkout pristine
171 # NOTE: Now, the tricky part (this is different from a real update)
172 git symbolic-ref HEAD refs/heads/master
173 # NOTE: Now, we think we're on the master branch, but we have
174 # pristine copy checked out
175 # NOTE: -p0 might need to be twiddled
176 patch -p0 < ../app-1.2.3/app-1.2.3.patch
178 # reconstitute .scripts directory
179 git checkout v1.2.2-scripts -- .scripts
181 # NOTE: Fake the merge
182 git rev-parse pristine > .git/MERGE_HEAD
184 You could also just try your luck with a manual merge using the patch
187 [ XXX: doc/layout.rst ]
188 * The repository for a given application will contain the following files:
190 - The actual application's files, as from the official tarball
192 - A .scripts directory, with the intent of holding Scripts specific files
193 if they become necessary.
195 - .scripts/dsn, overriding database source name