| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
| |
Before we were ignoring venv setup in the Makefile, so update the test runner
to use that instead of pestering developers to install those dependencies by
hand.
Issue #1632
|
|
|
|
|
|
| |
Ran 2to3 and fixed a few deprecated warnings
Issue #1632
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Force couch_replicator_auth_session plugin to refresh the session periodically.
Normally it is not needed as the session would be refreshed when requests start
failing with a 401 (authentication) or 403 (authorization) errors. In some
cases when anonymous writes are allowed to the database and a VDU function is
used to forbid writes based on the authenticated in username, requests with an
expired session cookie will not fail with a 401 and the session will not be
refreshed.
To fix the issue using these two approaches:
1. Use cookie's max-age expiry time to schedule a refresh. To ensure that time
is provided in the cookie, switch the the option to enable it by default. This
handles the issue for endpoints which are updated with this commit.
2. For endpoints which do not put a max-age time in the cookie, use a value
that's less than CouchDB's default auth timeout. If users changed their
auth timeout value, and use VDUs in the pattern described above, and don't
update their endpoints to version which sends max-age by default, they could
adjust `[replicator] session_refresh_interval_sec` to their auth timeout minus
some small delay.
Of course refresh based on auth/authz failures should still works as before.
Fixes #1607
|
| |
|
|\
| |
| | |
Fix couch_epi typespec for data provider
|
|/
|
|
|
|
|
|
|
|
| |
There were few problems:
- `module` was renamed into `static_module`
- https://github.com/apache/couchdb/commit/0fefc859eb9c18120064317da61a30adaeac5f92#diff-d9e3e3c91d4866fe966666619bda7991
- `callback_module` was added
- https://github.com/apache/couchdb/commit/cf65280466499d652cff1171a2039af49c5677e8#diff-d9e3e3c91d4866fe966666619bda7991
- data provider specification can include options
- https://github.com/apache/couchdb/blob/master/src/couch_epi/src/couch_epi_plugin.erl#L143
|
|\
| |
| | |
Do not use [] in feature_flags configuration
|
| | |
|
|/
|
|
|
|
|
|
| |
[] characters are fobidden in configuration keys for security reason.
Use list without [] to configure flags. Example configuration:
[feature_flags]
foo,bar,baz||test* = true
|
|
|
| |
Closes #1746
|
|\
| |
| |
| |
| | |
cloudant/84736-handle-db-deletion-in-load_validation_funs
handle db deletion in load validation funs
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously there were quite a few problems with load_validation_funs
in the case when clustered database is deleted.
- the calls to load_validation_funs were failing with `internal_server` error [1]
- the deleted database stayed opened because:
- the caller of the load_validation_funs (update_doc) stayed alive
- the main_pid of the deleted database wasn't killed either
- there was an infinite loop in ddoc_cache_entry trying to recover ddoc from deleted database
The solution is:
- do not call `recover` for deleted database
- close `main_pid`
- use `erlang:error` to crash the caller
[1] - The stack trace was:
```
{database_does_not_exist,[
{mem3_shards,load_shards_from_db,"bailey/meta",[
{file,"src/mem3_shards.erl"},{line,394}]},
{mem3_shards,load_shards_from_disk,1,[
{file,"src/mem3_shards.erl"},{line,369}]},
{mem3_shards,for_db,2,[
{file,"src/mem3_shards.erl"},{line,54}]},
{fabric_view_all_docs,go,5,[
{file,"src/fabric_view_all_docs.erl"},{line,24}]},
{ddoc_cache_entry_validation_funs,recover,1,[
{file,"src/ddoc_cache_entry_validation_funs.erl"},{line,33}]},
{ddoc_cache_entry,do_open,1,[
{file,"src/ddoc_cache_entry.erl"},{line,294}]}]}
```
|
|\
| |
| | |
Fix total_rows value for queries on `_design_docs` handler
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| | |
This adds a new fabric provider for getting a counter for design docs
in a database. It allows for a valid query for total_rows when _design_docs
handler queried with an array of keys.
|
|/ |
|
|
|
|
| |
extra field in the header.
|
|\
| |
| | |
Remove unnecessary setting on chttpd_view_test
|
|/ |
|
|\
| |
| | |
Add query server to eunit, Makefile.win
|
|/ |
|
| |
|
|\
| |
| | |
Avoid badmatch when getting fabric timeout
|
|/
|
|
|
|
|
| |
- Avoid badmatch when putting set_purge_infos_limit
and getting fabric timeout
COUCHDB-3326
|
|\
| |
| | |
Fix nasty 500 error on POST to /_config
|
|/ |
|
| |
|
|\
| |
| | |
Test suite fixes for Windows
|
| | |
|
| |
| |
| |
| |
| |
| | |
"It's the test suite, it can't be DNS!"
"It's ALWAYS DNS."
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unfortuantely, os:timestamp() on Windows only has millisecond
accuracy. That means that the two Mango tests checking for a positive
value of execution_time fail, since these tests appear to be running
in <1ms on my test setup (a rather anemic Windows VM!)
This change disables only the check for execution_time in two tests,
and leaves the remainder of the execution_stats checks in place
on Windows.
It also introduces a convenience "make.cmd" file so you can
"make check" without typing "make -f Makefile.win check" all the time.
|
| |
|
|\
| |
| | |
Feature flags
|
| | |
|
| | |
|
| | |
|
|/ |
|
|
|
|
| |
* Reformat mix.lock
* Suppress elixir 1.7 warnings
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Enable cluster auto-assembly through a seedlist
This introduces a new config setting which allows an administrator to
configure an initial list of nodes that should be contacted when a node
boots up:
[cluster]
seedlist = couchdb@node1.example.com,couchdb@node2.example.com,couchdb@node3.example.com
If configured, CouchDB will add every node in the seedlist to the _nodes
DB automatically, which will trigger a distributed Erlang connection and
a replication of the internal system databases to the local node. This
eliminates the need to explicitly add each node using the HTTP API.
We also modify the /_up endpoint to reflect the progress of the initial seeding
of the node. If a seedlist is configured the endpoint will return 404 until the
local node has updated its local replica of each of the system databases from
one of the members of the seedlist. Once the status flips to "ok" the endpoint
will return 200 and it's safe to direct requests to the new node.
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix bug in epoch millisecond calculation
* Include timeout milliseconds in exception message
* Support assertions in retry_until
Previously, retry_until would only work with boolean conditions.
This change supports assertions in code being retried, which requires
fewer changes to wrap existing code containing assertions.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously stats counts between job runs were reset. So if a job was stopped
and restarted by the scheduler, its docs_written, docs_read, doc_write_failures,
etc., counts would go back to 0. For doc_write_failures this was especially bad
as it hid the fact that some documents were not replicated to the target
because either a VDU failed or one of the limits were hit.
This change preserves stats across job runs. Everytime active tasks is updated,
the stats object in rep record of each job in scheduler's ets table will be
updated asynchronously. On next job start the job will reinitialize from last
saved stats.
Relates somewhat to issue #1159
|
| |
|
|
|
|
|
| |
The previous configuration `[daemons] httpsd = {chttpd, start_link, [https]}`
is also sill supported for backwards cmpatibility reasons.
|
| |
|