Re: Bug in tables column data in postgres database

From: Sarita Sharma <saritakumarisharma61(at)gmail(dot)com>
To: Michaeldba(at)sqlexec(dot)com
Cc: Steve Midgley <science(at)misuse(dot)org>, Erik Brandsberg <erik(at)heimdalldata(dot)com>, "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>, "Voillequin, Jean-Marc" <Jean-Marc(dot)Voillequin(at)moodys(dot)com>, pgsql-sql <pgsql-sql(at)lists(dot)postgresql(dot)org>
Subject: Re: Bug in tables column data in postgres database
Date: 2022-03-06 13:49:27
Message-ID: CAOKMkZpUAmNosfApAkiw4NjA+a6o6deRR6jaO9qMX9rHN5Ze-g@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql

Sorry, today l m not feeling good. Tomorrow I will reply you with all the
points you have asked me to do.

Thanks and Regards,
Sarita

On Sun, 6 Mar 2022, 17:53 Michaeldba(at)sqlexec(dot)com, <Michaeldba(at)sqlexec(dot)com>
wrote:

> You are not doing what we have asked you to do.
>
> I asked you to send me the ddl for the table including any triggers and
> also asked you to turn on debugging to capture dml changes
> (log_statement=mod) and you didn’t do it.
>
> You are not cooperating with all the folks that tried to help you.
>
> Sent from my iPad
>
> On Mar 5, 2022, at 6:14 PM, Sarita Sharma <saritakumarisharma61(at)gmail(dot)com>
> wrote:
>
> 
> Hello Team,
> I don't understand why Java layer is involved here. If data is saved in
> database table and with a select query wherever result will come that only
> getting fetched from java layers as a result set.
>
> If toggling behaviour is observed by me in tables column of postgres
> database then how Java layer involvement comes.
>
> At last my point of view is:
> Here I have reported something that seems to a bug as I observe some
> toggling behaviour . Just as a user only I have reported. And its all upto
> postres db team to take it into consideration or not instead of pointing
> out java layers.
>
> Upto this much responses I got I didn't got satisfied reply.
>
>
> Thankyou
>
>
>
> On Fri, 4 Mar 2022, 20:22 Steve Midgley, <science(at)misuse(dot)org> wrote:
>
>>
>>
>> On Fri, Mar 4, 2022, 6:33 AM Erik Brandsberg <erik(at)heimdalldata(dot)com>
>> wrote:
>>
>>> As others have said, it is nearly impossible for postgres to be usable
>>> in any environment with the bug you reported, and yet it is. It is 99.9999
>>> % likely the bug is somewhere else. You have provided no reproduction steps
>>> as well, so no action could even be taken if it were an issue. This is not
>>> a customer support channel with Salad as well. If you wish to have that
>>> type of support, you may wish to contact one of the many companies that
>>> provide support for postgresql.
>>>
>>> On Fri, Mar 4, 2022, 9:17 AM Sarita Sharma <
>>> saritakumarisharma61(at)gmail(dot)com> wrote:
>>>
>>>> Hello Team,
>>>> I am getting a genuine defect that from postgres database as I am
>>>> working for a well recognised organization and I am only handling coding
>>>> from layers that includes spring boot, Jpa configuration for database
>>>> connectivity and postgres as database.
>>>> The toggling issue I told , I have analysed it many times and I only
>>>> correcting data whenever my data got toggle like I have already explained.
>>>> If you find it a genuine issue , you can take it under consideration.
>>>> As a employee I have already informed this issue to my senior colleague.
>>>>
>>>> Thanks and Regards,
>>>> Sarita
>>>>
>>>> On Fri, 4 Mar 2022, 01:33 Steve Midgley, <science(at)misuse(dot)org> wrote:
>>>>
>>>>>
>>>>>
>>>>> On Thu, Mar 3, 2022 at 11:27 AM David G. Johnston <
>>>>> david(dot)g(dot)johnston(at)gmail(dot)com> wrote:
>>>>>
>>>>>> On Thu, Mar 3, 2022 at 12:13 PM Sarita Sharma <
>>>>>> saritakumarisharma61(at)gmail(dot)com> wrote:
>>>>>>
>>>>>>> I am working as a Java developer and for backend database I am using
>>>>>>> postgres database. I am generating this tables using entity classes of Java
>>>>>>> and Jpa configuration to handling database connectivity and table creation.
>>>>>>> I am finding this toggling issue on my day to day work activity . I
>>>>>>> am still working with same situation. And I only have to correct data
>>>>>>> whenever its value keeps changing automatically.
>>>>>>> Please take my concern .
>>>>>>>
>>>>>>>
>>>>>> You are going to have to produce a self-contained example program
>>>>>> (and database schema) that demonstrates the issue if you want to have any
>>>>>> hope of having it solved. Once you've done that, figuring out which piece
>>>>>> of software is "buggy" should become possible. Until you can manage that,
>>>>>> handing off the problem to someone to solve is not going to work.
>>>>>> David J.
>>>>>>
>>>>>
>>>>> Ditto this point, and to say that in my experience with databases in
>>>>> general and postgres specifically, I've never even heard of this type of
>>>>> problem manifesting, or similar to like it.
>>>>>
>>>>> Whereas, I've personally encountered this type of unexpected
>>>>> "toggling" of data in the ORM or other language-specific layers many, many
>>>>> times. So just from a parsimony perspective, it's wise to look at and
>>>>> instrument your java stack to find the problem: as it is almost certain
>>>>> that this type of problem is in those layers. If it isn't there, it could
>>>>> be in some kind of business logic code inside stored procedures in your
>>>>> postgres environment of course -- that's also a very viable path of
>>>>> inquiry. But the idea that postgres itself is toggling your data is so
>>>>> improbable, that you should discount it in your analysis. It's more likely
>>>>> you have flaky ram or a motherboard creating the issue (and those are also
>>>>> highly unlikely candidates for this problem).
>>>>> Steve
>>>>>
>>>>
>> I would add a little to this point. Whether it is postgres or it isn't,
>> the steps to debug are the same. The easiest first step is to instrument
>> your Java layer with something like New Relic and to start capturing your
>> postgres transaction log and feeding it to to an analysis environment of
>> some kind.
>>
>> Then start filtering for activity that will show changes to the data in
>> question. We all are confident this is not postgres core. But I am even
>> more confident it is not postgres core doing this silently with no evidence
>> in the transaction log. So look for activity in New Relic and pg
>> transactions that will show you when this is happening. It should be easier
>> to identify the issue then.
>>
>> I also on occasion work for well recognized companies - and I see
>> spectacular errors in those development processes. The difference is that
>> they have extensive quality processes to catch issues before they go to
>> production. Ideally you can also draw on those resources to help you here.
>>
>> Simplify and instrument your environment. Use analytic tools. Develop
>> steps to reproduce (and steps where the problem won't manifest. Let that
>> guide your problem identification. Seek help internally from those with
>> expertise of your product, the code, and the systems.
>>
>> Steve
>>
>>>

In response to

Browse pgsql-sql by date

  From Date Subject
Next Message Scott Macri 2022-03-07 01:59:31 ERROR: extra data after last expected column
Previous Message Michaeldba@sqlexec.com 2022-03-06 12:23:23 Re: Bug in tables column data in postgres database