| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Erik Rijkers <er(at)xs4all(dot)nl> |
| Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: select limit error in file_fdw |
| Date: | 2018-12-16 18:10:42 |
| Message-ID: | 25989.1544983842@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Erik Rijkers <er(at)xs4all(dot)nl> writes:
> On 2018-12-16 16:52, Tom Lane wrote:
>> However, something else occurred to me this morning, and a bit
>> later I can reproduce the problem! I did it by changing the
>> table's definition to use a shell pipeline:
> /bin/sh seems to be dash, here.
Hm. That must be the relevant difference. I just repeated the experiment
on a Fedora 28 box with reasonably up-to-date bash:
$ /bin/sh --version
GNU bash, version 4.4.23(1)-release (x86_64-redhat-linux-gnu)
and it behaves the same as my RHEL6 box: no problem with the direct
invocation of unzip, problem if use a pipeline.
Anyway, we know what to do, so I'll go do it.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tomas Vondra | 2018-12-16 18:32:22 | Re: reorderbuffer: memory overconsumption with medium-size subxacts |
| Previous Message | Tom Lane | 2018-12-16 18:05:03 | Why aren't we using strsignal(3) ? |