Summary:
Ref T1191. Currently if a developer forgot to specify a column type, `storage adjust` aborts explosively mid-stream. Instead:
- Make this a formal error with an unambiugous name/description instead of something you sort of infer by seeing "<unknown>".
- Make this error prevent generation of adjustment warnings, so we don't try to `ALTER TABLE t CHANGE COLUMN c <unknown>`, which is nonsense.
- When schemata errors exist, surface them prominiently in `storage adjust`.
Overall:
- Once `storage upgrade` runs `storage adjust` automatically (soon), this will make it relatively difficult to miss these errors.
- Letting these errors slip through no longer escalates into a more severe issue.
Test Plan:
Commented out the recent `mailKey` spec and ran `storage adjust`:
```
$ ./bin/storage adjust --force
Verifying database schemata...
Found no adjustments for schemata.
Target Error
phabricator2_phriction.phriction_document.mailKey Column Has No Specification
SCHEMATA ERRORS
The schemata have serious errors (detailed above) which the adjustment
workflow can not fix.
If you are not developing Phabricator itself, report this issue to the
upstream.
If you are developing Phabricator, these errors usually indicate that your
schema specifications do not agree with the schemata your code actually
builds.
```
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10771
Summary: Fixes T6436. We subclass the wrong controller and miss the admin-only check.
Test Plan: Ignored / unignored set up issues.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6436
Differential Revision: https://secure.phabricator.com/D10765
Summary:
Ref T1191. Notable stuff:
- Adds `--disable-utf8mb4` to `bin/storage` to make it easier to test what things will (approximately) do on old MySQL. This isn't 100% perfect but should catch all the major stuff. It basically makes us pretend the server is an old server.
- Require utf8mb4 to dump a quickstart.
- Fix some issues with quickstart generation, notably special casing the FULLTEXT handling.
- Add an `--unsafe` flag to `bin/storage adjust` to let it truncate data to fix schemata.
- Fix some old patches which don't work if the default table charset is utf8mb4.
Test Plan:
- Dumped a quickstart.
- Loaded the quickstart with utf8mb4.
- Loaded the quickstart with `--disable-utf8mb4` (verified that we get binary columns, etc).
- Adjusted schema with `--disable-utf8mb4` (got a long adjustment with binary columns, some truncation stuff with weird edge case test data).
- Adjusted schema with `--disable-utf8mb4 --unsafe` (got truncations and clean adjust).
- Adjusted schema back without `--disable-utf8mb4` (got a long adjustment with utf8mb4 columns, some invalid data on truncated utf8).
- Adjusted schema without `--disable-utf8mb4`, but with `--unsafe` (got truncations on the invalid data).
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10757
Summary: Fixes T4018. Basically hits the bullet points in that task description except the "ideally" one.
Test Plan:
ran bin/config migrate and saw sensible output.
```
~> ./bin/config migrate
Migrating file-based config to more modern config...
Skipping config of source type PhabricatorConfigDatabaseSource...
Skipping config of source type PhabricatorConfigLocalSource...
Skipping config of source type PhabricatorConfigDefaultSource...
Done. Migrated 0 keys.
```
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: hach-que, epriestley, Korvin
Maniphest Tasks: T4018
Differential Revision: https://secure.phabricator.com/D10490
Summary: Fixes T6119. This is a little fuzzy, but generally bumping up `innodb_buffer_pool_size` to something bigger than the default (which is often anemic, at `8M`) is desriable, and it seems like it will fix the specific issue a user encountered in T6119.
Test Plan: {F211855}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6119
Differential Revision: https://secure.phabricator.com/D10630
Summary:
Ref T1191. Although I fixed some of the mutations earlier (in D10598), I missed the column mutations under old versions of MySQL. In particular, this isn't valid:
- `ALTER TABLE ... MODIFY columnName VARCHAR(64) COLLATE binary`
Issue the permitted version of this instead, which is:
- `ALTER TABLE ... MODIFY columnName VARBINARY(64)`
Also fixed an issue where a clean schema had the wrong nullability for a column in the draft table. Force it to the expected nullability.
The other trick here is around the one column with a FULLTEXT index on it, which needs a little massaging.
Test Plan:
- Forced my local install to return `false` for utf8mb4 support.
- Did a clean adjust into `binary` columns.
- Poked around, added emoji to things.
- Reverted the fake check and did a clean adjust into `utf8mb4` columns.
- Emoji survived.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: fabe, epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10627
Summary:
Ref T1191. Now that the whole database is covered, we don't need to do as much work to build expected schemata. Doing them database-by-database was helpful in converting, but is just reudndant work now.
Instead of requiring every application to build its Lisk objects, just build all Lisk objects.
I removed `harbormaster.lisk_counter` because it is unused.
It would be nice to autogenerate edge schemata, too, but that's a little trickier.
Test Plan: Database setup issues are all green.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley, hach-que
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10620
Summary: Fixes T6210. The current messaging may be confusing if `pygmentize` is available but broken.
Test Plan: Faked the binary names and hit the errors, which seemed helpful.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6210
Differential Revision: https://secure.phabricator.com/D10626
Summary:
Ref T1191. When changing the column type of an AUTO_INCREMENT column, we currently may lose the autoincrement attribute.
Instead, support it. This is a bit messy because AUTO_INCREMENT columns interact with PRIMARY KEY columns (tables may only have one AUTO_INCREMENT column, and it must be a primary key). We need to migrate in more phases to avoid this issue.
Introduce new `auto` and `auto64` types to represent autoincrement IDs.
Test Plan:
- Saw autoincrement show up correctly in web UI.
- Fixed an autoincrement issue on the XHProf storage table with `bin/storage adjust` safely.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10607
Summary:
Ref T1191. Currently, the `quickstart.sql` gets generated in a pretty manual fashion. This is a pain, and will become more of a pain in the world of utf8mb4.
Provide a workflow which does upgrade + adjust + dump + destroy, then massages the output to produce a workable `quickstart.sql`.
Test Plan: Inspected output; I'll test this more throughly before actually generating a new quickstart, but that's some ways away.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10603
Summary:
Ref T1191. For most text columns, we either don't care if "a" and "A" are the same, or we expect them to be different (for example: keys, domains, secrets, etc). Default text columns to the `_bin` collation so they are compared by strict character value. This is safer in cases where we aren't sure.
For some text columns, we allow the user to sort by the column in the UI (like Maniphest task titles) or we do care that "A" and "a" are the same (for example: project names). Introduce a new class of virtual data types, the "sort..." types, to cover these columns. These are like the "text..." types but use sorting collations which treat "A" and "a" the same.
Test Plan:
- Made an effort to identify all columns where the UI relies on database collation.
- Ran `bin/storage adjust` and cleared all warnings.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: beng, epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10602
Summary:
Ref T1191. This was useful for annotating everything but we no longer need it; there are just two types of issues now:
- Error: stuff we can't fix (missing or surplus tables/database/columns, bad column nullability).
- Warning: stuff we can fix (column types, character sets, collations, missing or surplus keys, incorrectly defined keys, bad key uniqueness).
Test Plan: Saw 3,399 warnings and 0 errors.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10594
Summary:
Ref T1191. Ref T6203. While generating expected schemata, I ran into these columns which seem to have sketchy nullability.
- Mark most of them for later resolution (T6203). They work fine today and don't need to block T1191. Changing them can break the application, so we can't autofix them.
- Forgive a couple of them that are sort-of reasonable or going to get wiped out.
Test Plan: Saw 94 remaining warnings.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: hach-que, epriestley
Maniphest Tasks: T1191, T6203
Differential Revision: https://secure.phabricator.com/D10593
Summary:
Ref T1191. We have several keys on `<x, y, id>`. When `id` is an auto-increment primary key, I believe this is exactly equivalent to a key on `<x, y>`, because the leaf nodes are implicitly sorted by `id`. We omit the implicit `id` elsewhere.
It would be nice to drop the `id` bit for consistency, but it's not doing any harm and this doesn't need to block the primary work of T1191.
Test Plan: Saw slightly fewer warnings.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10592
Summary:
Ref T1191.
- Adds definitions for missing keys and keys with wrong uniqueness. Generally, I defined these before fixing the key query to actually pull all keys and support uniqueness.
- Moves "key uniqueness" to note severity; this is fixable (probably?) and there are no remaining issues.
- Moves "Missing Key" to note severity; missing keys are fixable and all remaining missing keys are really missing (either missing edge keys, or missing PHID keys):
{F210089}
- Moves "Surplus Key" to note seveirty; surplus keys are fixable all remaining surplus keys are really surplus (duplicate key in Harbormaster, key on unused column in Worker):
{F210090}
Test Plan:
- Vetted missing/surplus/unique messages.
- 146 issues remaining.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10590
Summary:
Ref T1191. Nothing too notable here:
- Allow a Lisk object to specify that there's no expectation that a table exists. We have one Harbormaster object and one Token object like this.
- Removed BuildPlanTransactionComment because it's currently unused.
Test Plan:
- Saw ~200 fewer warnings; just ~800 left.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10583
Summary:
Ref T1191.
- Removes ponder comment table. This was migrated a very long time ago.
Test Plan:
- Grepped for removed table.
- Saw ~100 fewer issues in web UI.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10582
Summary:
Ref T1191. Notes:
- Drops the project affiliation table. This is a very old membership table which was migrated to edges.
- Drops the subproject table. This is a very old table for a removed feature.
Test Plan:
- Grepped for dropped tables.
- Saw ~100 fewer setup issues.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10581
Summary:
Ref T1191. Some notes here:
- Drops the old LDAP and OAuth info tables. These were migrated to the ExternalAccount table a very long time ago.
- Separates surplus/missing keys from other types of surplus/missing things. In the long run, my plan is to have only two notice levels:
- Error: something we can't fix (missing database, table, or column; overlong key).
- Warning: something we can fix (surplus anything, missing key, bad column type, bad key columns, bad uniqueness, bad collation or charset).
- For now, retaining three levels is helpful in generating all the expected scheamta.
Test Plan:
- Saw ~200 issues resolve, leaving ~1,300.
- Grepped for removed tables.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10580
Summary: Ref T6185. Although it seems that we can't easily defuse or mitigate this, we can at least warn administrators.
Test Plan: Ran on my (unpatched, local) system, got a setup warning.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T6185
Differential Revision: https://secure.phabricator.com/D10561
Summary: Ref T1191. Handful of minor things here (T6150, T6149, T6148, T6147, T6146) but nothing very noteworthy.
Test Plan: Viewed web UI, saw fewer errors.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10527
Summary:
Ref T1191.
- Adds support for custom fields.
- Adds support for partial indexes (indexes on a prefix of a column).
- Drops old auxiliary storage table: this was moved to custom field storage about a year ago.
- Drops old project table: this was moved to edges about two months ago.
Test Plan:
- Viewed web UI, saw fewer issues.
- Used `grep` to verify no readers/writers for storage or project table.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10526
Summary:
Ref T1191. Three parts:
- The old way of getting key information only got primary / unique / foreign keys, not all keys. Use `SHOW INDEXES` to get all keys instead.
- Track key uniqueness and raise warnings about it.
- Add a new "all issues" view to show an expanded, flat view of all issues. This is just an easier way to get a list so you don't have to dig around in the hierarchical view.
Test Plan:
{F206351}
{F206352}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10525
Summary:
Ref T1191.
- There was a varchar(50) column. I changed it to `text64`, since this length is unusual.
- There was an int(3) column. I changed it to `int32`, since this length is unusual.
Test Plan: Ran migrations, saw warnings disappear from config tool.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10524
Summary: T1191. Nothing very notable here.
Test Plan: Saw more blue in web UI.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D10522
Summary:
Ref T1191. Notable:
- Allowed objects to remove default columns (some feed tables have no `id`).
- Added a "note" severity and moved all the charset stuff down to that to make progress more clear.
Test Plan:
Trying to make the whole thing blue...
{F205970}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10519
Summary: Ref T1191. Fills in some more of the databases. Nothing very notable here. I didn't encounter any issues or overlong keys.
Test Plan: Used web UI to click around and verify expected schemata match up against actual schemata well.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10516
Summary: Ref T1191. This fills in some more features and gets audit and auth nearly generating reasonable expected schemata.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10500
Summary:
Ref T1191. The major issue motivation here is that InnoDB keys have a maximum length of 767 bytes. When we move `utf8` colums to `utf8mb4` columns, they'll jump from 3 bytes per character to 4 bytes per character, which may make some indexes too long. Add key schema to help spot this.
Also add nullability since it doesn't hurt.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10499
Summary:
Ref T1191. This lays some groundwork for generating the expected schemata, so we can compare them to the actual schemata and produce a meaningful diff.
- In general, each application will subclass `PhabricatorConfigSchemaSpec` and provide a definition of the tables it expects.
- This class has helper methods to mostly-automatically build table definitions for Lisk and (in the future) edges.
- When building expected schema, we specify a "data type", like "epoch". This is the type of data the application stores in the column, from the application's point of view. The SchemaSpec converts this into the best avilable storage type: for example, "text" will translate to `utf8mb4` if it's availalbe, or `binary` if not. This gives us a layer of indirection to insulate us from craziness.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10497
Summary:
Ref T1191. This builds on the "view of the database as it exists" by building a view of the database as it is expected to exist (this is mostly empty for now) and comparing the two. We now render a view of the "comparison schema", which is the actual schema merged with the expected schema and annotated with the differences.
(I'm merging them like this because it makes it easier to handle both "missing" and "surpulus" warnings in a consistent way. If we tried to annotate just the actual or expected schema, the absence of components which are expected to exist is messy to handle.)
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10496
Summary:
Ref T1191. Plan here is:
- Build a tool showing the current schemata status (this diff).
- Have it compare the current status to the desired status (partly here, mostly in future diffs).
- Then add a migration tool, and eventually a setup issue to tell people to run it.
Test Plan:
Reviewed current schemata.
{F204492}
{F204493}
{F204494}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10494
Summary: See rP8806fb0296c2.
Test Plan:
me fail english
with bonus!
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D10514
Summary:
Fixes T6084. Changes:
- Rename `phabricator.show-beta-applications` to `phabricator.show-prototypes`, to reinforce that these include early-development applications.
- Migrate the config setting.
- Add an explicit "no support" banner to the config page.
- Rename "Beta" to "Prototype" in the UI.
- Use "bomb" icon instead of "half star" icon.
- Document prototype applications in more detail.
- Explicitly document that we do not support these applications.
Test Plan:
- Ran migration.
- Resolved "obsolete config" issue.
- Viewed config setting.
- Browsed prototypes in Applications app.
- Viewed documentation.
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: epriestley, hach-que
Maniphest Tasks: T6084
Differential Revision: https://secure.phabricator.com/D10493
Summary:
Fixes T6044. We've had two cases (both the same install, coincidentally) where pages got hung doing too much data fetching.
When pages hang, we don't get a useful stack trace out of them, since nginx, php-fpm, or PHP eventually terminates things in a non-useful way without any diagnostic information.
The second time (the recent Macros issue) I was able to walk the install through removing limits on nginx, php-fpm, php, and eventually getting a profile by letting the page run for several minutes until the request completed. However, this install is exceptionally technically proficient and this was still a big pain for everyone, and this approach would not have worked if the page actually looped rather than just taking a long time.
Provide `debug.time-limit`, which should give us a better tool for reacting to this situation: by setting it to a small value (like 10), we'll kill the page after 10 seconds with a trace, before nginx/php-fpm/php/etc can kill it uselessly. Hopefully that will be enough information to find the issue (generally, getting a trace has been 95% of the problem in the two cases we've encountered).
Test Plan: Set this option to `3` and added a sleep loop, saw a termination after 3 seconds with a useful trace.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: csilvers, joshuaspence, epriestley
Maniphest Tasks: T6044
Differential Revision: https://secure.phabricator.com/D10465
Summary: Fixes T6052. Allow installs to link to legal documents, etc., in the page footer.
Test Plan:
- Configured a footer.
- Viewed workboards (no footer).
- Viewed Conpherence (no apparent disruption, I think everything z-indexes over the footer).
- Viewed stuff on mobile (seems OK).
- Viewed login page (saw footer).
{F201718}
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T6052
Differential Revision: https://secure.phabricator.com/D10466
Summary:
Ref T5968. Issues we've seen from users include:
- Concern about severity ("... Need Restarting").
- Reduce severity of explanatory text ("Different Config", "not severe").
- Explain consequences in more detail.
- In D10420, make "Ignore" easier to find.
- Scope language for the multi-machine case ("at least one daemon").
- Confusion about why daemons need restarting.
- Unbury the lede ("Daemons and Web Have Different Config").
- Make it clear that the root cause is a different checksum by showing the checksum. (This just hammers home that we're comparing checksums and this issue is about config checksums and we're not making it up, the checksums probably aren't that useful on their own.)
- Difficulty understanding how to proceed when restarting does not resolve the issue:
- Call out steps to take on the daemon console explicitly.
- Walk through troubleshooting PHABRICATOR_ENV.
- Walk through troubleshooting multiple `local.json`.
Test Plan: {F199245}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5968
Differential Revision: https://secure.phabricator.com/D10421
Summary:
Ref T4331. Ref T5968. Users sometimes have trouble figuring out how to ignore issues. The option is a bit hard to spot, especially if you aren't familiar with interfaces yet.
Make it a button on the issue page itself instead.
Test Plan:
Normal issue:
{F199225}
Ignored issue:
{F199226}
Fatal issue:
{F199227}
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T4331, T5968
Differential Revision: https://secure.phabricator.com/D10420
Summary: Fixes T6011. See that task for discussion. We can detect when `memory_limit` will be the limiting factor for drag-and-drop uploads and warn administrators about it.
Test Plan: Fiddled configuration values and hit, then resolved, the issue.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6011
Differential Revision: https://secure.phabricator.com/D10413
