You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@beam.apache.org by Pablo Estrada <pa...@google.com> on 2019/05/08 21:35:34 UTC

[discuss] Reducing log verbosity for Python failures?

Hello all,
Some tests in Python have the problem that when they fail, lots of internal
logging is dumped onto stdout, and we end up having to scroll way up to
find the actual stack trace for the failed test. This logging, as far as i
can tell, is dumping of fn api protos.

Does anyone use these logs to look into the test failure? I would like to
find a way to make these more compact, or maybe just stop logging them
(people who need them can choose to log them in their local setup?).

I lean towards making them more compact (by, for instance, writing
functions that log their information in a more compact fashion); but I
would like to hear thoughts from others.

So thoughts? : )
-P.

Re: [discuss] Reducing log verbosity for Python failures?

Posted by Ahmet Altay <al...@google.com>.
In that case compaction makes much more sense. My experience was when a
single test in a suite fails, I end up seeing lots of logs (usually proto
dumps) from unrelated tests.

*From: *Pablo Estrada <pa...@google.com>
*Date: *Wed, May 8, 2019 at 3:48 PM
*To: *dev

My impression - and I might be wrong [see meme at 1], is that these are
> logged in debug mode, but when the test fails, the test runner dumps them
> all to stdout.
> Best
> -P.
>
> [1]
> https://i1.wp.com/gifrific.com/wp-content/uploads/2015/02/Chris-Farley-Oh-God-Theyre-Gonna-Know-Im-Dumb-Conan-Interview.gif?ssl=1
>
> *From: *Ahmet Altay <al...@google.com>
> *Date: *Wed, May 8, 2019 at 3:13 PM
> *To: *dev
>
> +1 It is hard to debug with lots logs messages. And if anybody is using
>> them for development we can make those logs debug logs and hide them by
>> default.
>>
>> *From: *Robert Bradshaw <ro...@google.com>
>> *Date: *Wed, May 8, 2019 at 3:01 PM
>> *To: *dev
>>
>> +1 to making them significantly more compact in most cases.
>>>
>>> From: Pablo Estrada <pa...@google.com>
>>> Date: Wed, May 8, 2019 at 11:35 PM
>>> To: dev
>>>
>>> > Hello all,
>>> > Some tests in Python have the problem that when they fail, lots of
>>> internal logging is dumped onto stdout, and we end up having to scroll way
>>> up to find the actual stack trace for the failed test. This logging, as far
>>> as i can tell, is dumping of fn api protos.
>>> >
>>> > Does anyone use these logs to look into the test failure? I would like
>>> to find a way to make these more compact, or maybe just stop logging them
>>> (people who need them can choose to log them in their local setup?).
>>> >
>>> > I lean towards making them more compact (by, for instance, writing
>>> functions that log their information in a more compact fashion); but I
>>> would like to hear thoughts from others.
>>> >
>>> > So thoughts? : )
>>> > -P.
>>>
>>

Re: [discuss] Reducing log verbosity for Python failures?

Posted by Pablo Estrada <pa...@google.com>.
My impression - and I might be wrong [see meme at 1], is that these are
logged in debug mode, but when the test fails, the test runner dumps them
all to stdout.
Best
-P.

[1]
https://i1.wp.com/gifrific.com/wp-content/uploads/2015/02/Chris-Farley-Oh-God-Theyre-Gonna-Know-Im-Dumb-Conan-Interview.gif?ssl=1

*From: *Ahmet Altay <al...@google.com>
*Date: *Wed, May 8, 2019 at 3:13 PM
*To: *dev

+1 It is hard to debug with lots logs messages. And if anybody is using
> them for development we can make those logs debug logs and hide them by
> default.
>
> *From: *Robert Bradshaw <ro...@google.com>
> *Date: *Wed, May 8, 2019 at 3:01 PM
> *To: *dev
>
> +1 to making them significantly more compact in most cases.
>>
>> From: Pablo Estrada <pa...@google.com>
>> Date: Wed, May 8, 2019 at 11:35 PM
>> To: dev
>>
>> > Hello all,
>> > Some tests in Python have the problem that when they fail, lots of
>> internal logging is dumped onto stdout, and we end up having to scroll way
>> up to find the actual stack trace for the failed test. This logging, as far
>> as i can tell, is dumping of fn api protos.
>> >
>> > Does anyone use these logs to look into the test failure? I would like
>> to find a way to make these more compact, or maybe just stop logging them
>> (people who need them can choose to log them in their local setup?).
>> >
>> > I lean towards making them more compact (by, for instance, writing
>> functions that log their information in a more compact fashion); but I
>> would like to hear thoughts from others.
>> >
>> > So thoughts? : )
>> > -P.
>>
>

Re: [discuss] Reducing log verbosity for Python failures?

Posted by Ahmet Altay <al...@google.com>.
+1 It is hard to debug with lots logs messages. And if anybody is using
them for development we can make those logs debug logs and hide them by
default.

*From: *Robert Bradshaw <ro...@google.com>
*Date: *Wed, May 8, 2019 at 3:01 PM
*To: *dev

+1 to making them significantly more compact in most cases.
>
> From: Pablo Estrada <pa...@google.com>
> Date: Wed, May 8, 2019 at 11:35 PM
> To: dev
>
> > Hello all,
> > Some tests in Python have the problem that when they fail, lots of
> internal logging is dumped onto stdout, and we end up having to scroll way
> up to find the actual stack trace for the failed test. This logging, as far
> as i can tell, is dumping of fn api protos.
> >
> > Does anyone use these logs to look into the test failure? I would like
> to find a way to make these more compact, or maybe just stop logging them
> (people who need them can choose to log them in their local setup?).
> >
> > I lean towards making them more compact (by, for instance, writing
> functions that log their information in a more compact fashion); but I
> would like to hear thoughts from others.
> >
> > So thoughts? : )
> > -P.
>

Re: [discuss] Reducing log verbosity for Python failures?

Posted by Robert Bradshaw <ro...@google.com>.
+1 to making them significantly more compact in most cases.

From: Pablo Estrada <pa...@google.com>
Date: Wed, May 8, 2019 at 11:35 PM
To: dev

> Hello all,
> Some tests in Python have the problem that when they fail, lots of internal logging is dumped onto stdout, and we end up having to scroll way up to find the actual stack trace for the failed test. This logging, as far as i can tell, is dumping of fn api protos.
>
> Does anyone use these logs to look into the test failure? I would like to find a way to make these more compact, or maybe just stop logging them (people who need them can choose to log them in their local setup?).
>
> I lean towards making them more compact (by, for instance, writing functions that log their information in a more compact fashion); but I would like to hear thoughts from others.
>
> So thoughts? : )
> -P.