Conversation
|
We also need to add a test to catch the bug/fix that was introduced in 2.1.19 which was what caused all those issue back then. |
|
@drew-gross you remember what the problem was? Reconnection issues? |
|
It might have been related to this? https://github.com/christkv/mongodb-core/issues/112 |
|
YEs seems like it, I need to drill down further to make sure our mongo adapter properly reconnects after losing the connection. |
|
@flovilmart updated the pull request. |
|
Tested with very short socketTimoutMS values and it's magic. |
- Just because this way we can let the adapter an opportunity to yield a new error.... OR A SUCCESS
flovilmart
force-pushed
the
mongodb-client-2.2.4
branch
from
377d5c4 to
b3671df
Compare
|
@flovilmart updated the pull request. |
spec/index.spec.js
Outdated
| expect(body.code).toEqual(1); | ||
| expect(body.message).toEqual('Internal server error.'); | ||
| // Reconfigure with null so the next beforeEach don't fail trying to delete | ||
| reconfigureServer({ databaseAdapter: null }); |
There was a problem hiding this comment.
Use reconfigureServer({ databaseAdapter: null }).then(done); please, reconfigureServer returns a promise.
|
@flovilmart updated the pull request. |
|
@flovilmart updated the pull request. |
* chore(package): update mongodb to version 2.2.4 https://greenkeeper.io/ * fix for mongo 2.2.4 * Delete schema promise on error... - Just because this way we can let the adapter an opportunity to yield a new error.... OR A SUCCESS * Fixes test * Restore fix
* chore(package): update mongodb to version 2.2.4 https://greenkeeper.io/ * fix for mongo 2.2.4 * Delete schema promise on error... - Just because this way we can let the adapter an opportunity to yield a new error.... OR A SUCCESS * Fixes test * Restore fix
5 participants
We needed to tweak a bit the test so the next beforeEach won't crash trying to delete all data with a broken connection as the new mongo client will fail.