Review Board 3.0.13 is released

[Christian Hammond]

Hi everyone!

We released Review Board 3.0.13 today, which has a handful of behind-the-scenes performance improvements, some new features (the Follow menu, better control over e-mail sending to avoid "this is a suspicious e-mail" warnings, and a few others), and some bug fixes (including a regression when clearing the Target Groups field).

See the official announcement and full release notes for all the details and some screenshots.

Christian

--

Christian Hammond

President/CEO of Beanbag

Makers of Review Board

[André Klitzing]

Hi,

looks like our review bots are disabled now with 3.0.13. We upgraded from 3.0.12. Since we upgraded to 3.0.13 no review bot is started. Even the bot with condition "Always match" won't do anything.

That's really critical here. :-(

Best regards

   André Klitzing

[Christian Hammond]

Hi André,

When you say disabled, do you mean they're no longer handling requests? We haven't seen this ourselves, and I don't know why it'd suddenly be happening. The only things that touch integrations have to do with the configuration forms, and wouldn't impact existing configurations.

Does downgrading to 3.0.12 fix this for you?

Christian

--
Supercharge your Review Board with Power Pack: https://www.reviewboard.org/powerpack/
Want us to host Review Board for you? Check out RBCommons: https://rbcommons.com/
Happy user? Let us know! https://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups "Review Board Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[André Klitzing]

Hi Christian,

I forget to look into the log of reviewboard itself. Looks like this is the issue:

 root - Unable to load bad condition set data for integration configuration ID=25 for key="conditions"
Traceback (most recent call last):
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/reviewboard/integrations/models.py", line 68, in load_conditions
    'local_site': self.local_site,
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/djblets/conditions/conditions.py", line 316, in deserialize
    for i, condition_data in enumerate(data.get('conditions', []))
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/djblets/conditions/conditions.py", line 139, in deserialize
    condition_index=condition_index)
InvalidConditionValueError: A value is required.

I already opened all configurations and pressed "save" again.

Best regards
   André Klitzing

[André Klitzing]

Another error some day ago:

 - root - Unable to load bad condition set data for integration configuration ID=13 for key="conditions"
Traceback (most recent call last):
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/reviewboard/integrations/models.py", line 68, in load_conditions
    'local_site': self.local_site,
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/djblets/conditions/conditions.py", line 316, in deserialize
    for i, condition_data in enumerate(data.get('conditions', []))
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/djblets/conditions/conditions.py", line 131, in deserialize
    value = operator.value_field.deserialize_value(raw_value)
  File "/opt/reviewboard/dist/lib/python2.7/site-packages/djblets/conditions/values.py", line 271, in deserialize_value
    code=e.code)
InvalidConditionValueError: Select a valid choice. 4 is not one of the available choices.

[Christian Hammond]

Hi André,

Are you making use of Local Site functionality?

Christian

--
Supercharge your Review Board with Power Pack: https://www.reviewboard.org/powerpack/
Want us to host Review Board for you? Check out RBCommons: https://rbcommons.com/
Happy user? Let us know! https://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups "Review Board Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Christian Hammond]

I'm probably going to need copies of the settings for one of the integrations in order to properly diagnose this, along with knowing about the Local Site usage. We did make some changes that prevent conditions from spanning Local Sites (it was a bug that an integration configuration not on a Local Site could interface with data on other Local Sites, or vice-versa). If you're not using Local Sites, though, then that's pretty concerning, and something I want to fix fast, but I'm not hitting it here.

Christian

[A. Klitzing]

Hi Christian,

No, we do not use it.

Maybe it is enabled but unused?

You received this message because you are subscribed to a topic in the Google Groups "Review Board Community" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/reviewboard/BL6t2et3f9o/unsubscribe.
To unsubscribe from this group and all its topics, send an email to reviewboard...@googlegroups.com.

[Christian Hammond]

They'd be explicitly configured, with things like review groups and repositories being bound to them. You'd have to go out of your way to use it, so probably not.

Are you using private repositories or invite-only groups with these configurations?

Can you show me the configuration for the integration shown in the log above (ID 13)?

Christian

[André Klitzing]

No, we do not have any private repositories and no invite-only groups.

Here is the config from database screen:

{
  "comment_on_unmodified_code": false,
  "conditions": {
    "conditions": [
      {
        "choice": "review-groups",
        "op": "contains-any",
        "value": [
          5,
          1,
          11,
          7,
          15
        ]
      },
      {
        "choice": "repository",
        "op": "one-of",
        "value": [
          1
        ]
      }
    ],
    "mode": "any"
  },
  "max_comments": 30,
  "open_issues": true,
  "tool": 10,
  "tool_options": "{\"max_line_length\": \"120\", \"encoding\": \"utf-8\"}"
}

[Christian Hammond]

Which error are you getting when saving that entry?

Do any of those review groups have the "Visible" checkbox unchecked? There was an update to the query used for review groups validation that resulted in hidden groups being excluded from results.

Christian

--
Supercharge your Review Board with Power Pack: https://www.reviewboard.org/powerpack/
Want us to host Review Board for you? Check out RBCommons: https://rbcommons.com/
Happy user? Let us know! https://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups "Review Board Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[André Klitzing]

I don't get any error if I save the entry.

Any review group is visible.

[André Klitzing]

I changed logging to "DEBUG" and posted a test review request. Here is the log.

[Christian Hammond]

Thanks. I'll dig into this some more, try to reproduce. For now, I recommend downgrading Review Board and Djblets.

Christian

On Tue, Feb 26, 2019 at 2:55 PM André Klitzing <akli...@gmail.com> wrote:

I changed logging to "DEBUG" and posted a test review request. Here is the log.

--

Supercharge your Review Board with Power Pack: https://www.reviewboard.org/powerpack/
Want us to host Review Board for you? Check out RBCommons: https://rbcommons.com/
Happy user? Let us know! https://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups "Review Board Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Christian Hammond]

I've found the cause. We'll get this fixed for the next release.

Christian

[André Klitzing]

Thanks! It works for me....

By the way: If I download the patch with the link the content is duplicated:

https://reviews.reviewboard.org/r/10435/diff/raw/