Обсуждение: Returning a memory issue when accessing an attachment from themailing list

Hi,

When trying to download the following attachment from the
pgsql-hackers mailing list archive I get the following message:
<memory at 0x7fa951355708>

The attachment is:
https://www.postgresql.org/message-id/attachment/108448/v2-0001-Improve-handling-of-parameter-differences-in-phys.patch

Kind regards,

--
Martín Marqués                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

Hi Martin,

On 4/2/20 6:11 PM, Martín Marqués wrote:
> Hi,
>
> When trying to download the following attachment from the
> pgsql-hackers mailing list archive I get the following message:
> <memory at 0x7fa951355708>
>
> The attachment is:
>
https://www.postgresql.org/message-id/attachment/108448/v2-0001-Improve-handling-of-parameter-differences-in-phys.patch

Can you please provide a URL to the exact message as well? A quick test
of some random messages did not trigger the same issue.

Thanks,

Jonathan

Hi,

> > When trying to download the following attachment from the
> > pgsql-hackers mailing list archive I get the following message:
> > <memory at 0x7fa951355708>
> >
> > The attachment is:
> >
https://www.postgresql.org/message-id/attachment/108448/v2-0001-Improve-handling-of-parameter-differences-in-phys.patch
>
> Can you please provide a URL to the exact message as well? A quick test
> of some random messages did not trigger the same issue.

https://www.postgresql.org/message-id/flat/4ad69a4c-cc9b-0dfe-0352-8b1b0cd36c7b@2ndquadrant.com

Kind regards,

--
Martín Marqués                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

On 4/2/20 6:45 PM, Martín Marqués wrote:
> Hi,
>
>>> When trying to download the following attachment from the
>>> pgsql-hackers mailing list archive I get the following message:
>>> <memory at 0x7fa951355708>
>>>
>>> The attachment is:
>>>
https://www.postgresql.org/message-id/attachment/108448/v2-0001-Improve-handling-of-parameter-differences-in-phys.patch
>>
>> Can you please provide a URL to the exact message as well? A quick test
>> of some random messages did not trigger the same issue.
>
> https://www.postgresql.org/message-id/flat/4ad69a4c-cc9b-0dfe-0352-8b1b0cd36c7b@2ndquadrant.com

Thanks.

I don't have access to the backend, but looking at the code my best
guess is that the patch did not import properly. The import script just
needs to be run against it again.

I downloaded the mbox file and it appears that the patch is in there. As
a workaround, if you download the mbox file and import it into your mail
client, you should be able to view and download it.

Thanks,

Jonathan

Hi,

> > https://www.postgresql.org/message-id/flat/4ad69a4c-cc9b-0dfe-0352-8b1b0cd36c7b@2ndquadrant.com
>
> Thanks.
>
> I don't have access to the backend, but looking at the code my best
> guess is that the patch did not import properly. The import script just
> needs to be run against it again.
>
> I downloaded the mbox file and it appears that the patch is in there. As
> a workaround, if you download the mbox file and import it into your mail
> client, you should be able to view and download it.

I was able to read the patch from the git repository, as it was
already committed in the master branch. I wanted to point out the
issue in case there was something not working. Seems it was just an
isolated issue with this attachment.

Thank you very much for looking into it.

Kind regards,

--
Martín Marqués                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

=?UTF-8?B?TWFydMOtbiBNYXJxdcOpcw==?= <martin@2ndquadrant.com> writes:
>> I downloaded the mbox file and it appears that the patch is in there. As
>> a workaround, if you download the mbox file and import it into your mail
>> client, you should be able to view and download it.

> I was able to read the patch from the git repository, as it was
> already committed in the master branch. I wanted to point out the
> issue in case there was something not working. Seems it was just an
> isolated issue with this attachment.

It's not an isolated issue: a patch I just posted at

https://www.postgresql.org/message-id/7653.1585869509@sss.pgh.pa.us

is having the same problem.  If you click on the link embedded on
that page,

https://www.postgresql.org/message-id/attachment/109592/gin-fuzzy-search-limit-fix-2.patch

all you see is

<memory at 0x7fa950fcf1c8>

The cfbot is also seeing that, apparently. because it's complaining
about a malformed patch --- but the copy that came back to me from
the mailing list is fine.

            regards, tom lane

Hi,

> It's not an isolated issue: a patch I just posted at
>
> https://www.postgresql.org/message-id/7653.1585869509@sss.pgh.pa.us
>
> is having the same problem.  If you click on the link embedded on
> that page,
>
> https://www.postgresql.org/message-id/attachment/109592/gin-fuzzy-search-limit-fix-2.patch

I'm able to download this patch perfectly. But the one I brought up is
still giving the memory message I reported. Maybe there's a process
still working that is fixing all the attachment issues?

Thanks,

--
Martín Marqués                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

Hi,

On 2020-04-02 21:59:27 -0300, Martín Marqués wrote:
> > It's not an isolated issue: a patch I just posted at
> >
> > https://www.postgresql.org/message-id/7653.1585869509@sss.pgh.pa.us
> >
> > is having the same problem.  If you click on the link embedded on
> > that page,
> >
> > https://www.postgresql.org/message-id/attachment/109592/gin-fuzzy-search-limit-fix-2.patch

If I had to guess the "<memory at 0x7fa950fcf1c8>" is caused by part of
the stack having been upgraded/downgraded. With the memoryview type not
being recognized with whatever is supposed to write out the data.


> Looks like t I'm able to download this patch perfectly.

Did you check whether the contents are correct? It downloads for me, but
it contains "<memory at 0x7fa950fcf1c8>".

Greetings,

Andres Freund

Greetings,

* Andres Freund (andres@anarazel.de) wrote:
> On 2020-04-02 21:59:27 -0300, Martín Marqués wrote:
> > > It's not an isolated issue: a patch I just posted at
> > >
> > > https://www.postgresql.org/message-id/7653.1585869509@sss.pgh.pa.us
> > >
> > > is having the same problem.  If you click on the link embedded on
> > > that page,
> > >
> > > https://www.postgresql.org/message-id/attachment/109592/gin-fuzzy-search-limit-fix-2.patch
>
> If I had to guess the "<memory at 0x7fa950fcf1c8>" is caused by part of
> the stack having been upgraded/downgraded. With the memoryview type not
> being recognized with whatever is supposed to write out the data.
>
> > Looks like t I'm able to download this patch perfectly.
>
> Did you check whether the contents are correct? It downloads for me, but
> it contains "<memory at 0x7fa950fcf1c8>".

For the archives sake, just wanted to mention that this has been
resolved.  If you see anything like this again, please don't hesitate to
reach out.

Thanks,

Stephen