public inbox for [email protected]
help / color / mirror / Atom feedFrom: Nikhil Shetty <[email protected]>
To: Adelino Silva <[email protected]>
Cc: Pgsql-admin <[email protected]>
Subject: Re: Postgres Stale Statistics
Date: Tue, 10 May 2022 23:44:02 +0530
Message-ID: <CAFpL5VyuYPS9on8ANA8vQ4Fdr18NWGvdGcD2YnB-ZdVFtPHwvg@mail.gmail.com> (raw)
In-Reply-To: <CAFpL5VxtUXwGSAU4200drsaqkv+pN206Z0RKT1dX=e8=C4Ls7w@mail.gmail.com>
References: <CAFpL5VzjDi35wo3+P19+kc7wGPh5=Xc-kCUfNDoTg3xNrcHnCw@mail.gmail.com>
<SA0PR15MB3856E1C18660DE3F01942DAED2FA9@SA0PR15MB3856.namprd15.prod.outlook.com>
<CAFpL5VzrX4To00Qv8xyQJsaJ+0gc7y4bY7pUWbg7Ffh47ZVuYA@mail.gmail.com>
<SA0PR15MB3856D4DBB389A9DF158433BDD2FA9@SA0PR15MB3856.namprd15.prod.outlook.com>
<CAFpL5VxgMsRTwvvdS2WFCbmoZspLwNxWOTHMQ089ou5SjP4J8g@mail.gmail.com>
<SA0PR15MB38568AF0F1533CF282F36F8BD2FD9@SA0PR15MB3856.namprd15.prod.outlook.com>
<CAFpL5VxtUXwGSAU4200drsaqkv+pN206Z0RKT1dX=e8=C4Ls7w@mail.gmail.com>
Hi,
Any inputs on how we can debug this further?
Thanks,
Nikhil
On Sun, 8 May 2022 at 12:12 PM, Nikhil Shetty <[email protected]>
wrote:
> Hi Adelino,
>
> About the EAGAIN (Resource temporarily unavailable).
>> UDP is a stateless protocol, unlike TCP which is connection oriented. The
>> recvfrom() code will not know whether or not the sender has closed its
>> socket, it only knows whether or not there is data waiting to be read.
>> According to the man page for recvfrom on Linux:
>> If no messages are available at the socket, the receive calls wait
>> for a message to arrive, unless the socket is nonblocking (see fcntl(2)) in
>> which case the value -1 is returned and the external variable errno set to
>> EAGAIN.
>
>
> Are you saying the message '-1 EAGAIN (Resource temporarily unavailable)'
> is normal?
>
> may you need to explore other option like disk saturation.
>> using stale statistics instead of current ones because stats collector is
>> not responding
>> <https://opensourcedbtech.com/2018/04/03/using-stale-statistics-instead-of-current-ones-because-stats...;
>
>
> There is no disk saturation from what we see.
>
> Thanks,
> Nikhil
>
> On Thu, Apr 28, 2022 at 2:52 PM Adelino Silva <[email protected]>
> wrote:
>
>> Hi Nikhil,
>>
>> About the EAGAIN (Resource temporarily unavailable).
>> UDP is a stateless protocol, unlike TCP which is connection oriented. The
>> recvfrom() code will not know whether or not the sender has closed its
>> socket, it only knows whether or not there is data waiting to be read.
>> According to the man page for recvfrom on Linux:
>>
>> If no messages are available at the socket, the receive calls wait
>> for a message to arrive, unless the socket is nonblocking (see fcntl(2)) in
>> which case the value -1 is returned and the external variable errno set to
>> EAGAIN.
>>
>>
>> may you need to explore other option like disk saturation.
>> using stale statistics instead of current ones because stats collector is
>> not responding
>> <https://opensourcedbtech.com/2018/04/03/using-stale-statistics-instead-of-current-ones-because-stats...;
>>
>> Regards,
>>
>> Adelino Silva
>> ------------------------------
>> *From:* Nikhil Shetty <[email protected]>
>> *Sent:* Wednesday, April 27, 2022 4:36 PM
>> *To:* Adelino Silva <[email protected]>
>> *Cc:* Pgsql-admin <[email protected]>
>> *Subject:* [EXTERNAL] Re: Postgres Stale Statistics
>>
>> Hi Adelino, I went through the article and I see there is no issue with
>> IPv6 in our case, it is using IPv4. I used strace and found 'Resource
>> temporarily unavailable' error though, not sure what this means, does this
>> mean there is an
>> ZjQcmQRYFpfptBannerStart
>> This Message Is From an External Sender
>> This message came from outside your organization.
>>
>> ZjQcmQRYFpfptBannerEnd
>> Hi Adelino,
>>
>> I went through the article and I see there is no issue with IPv6 in our
>> case, it is using IPv4.
>>
>>
>> I used strace and found 'Resource temporarily unavailable' error though,
>> not sure what this means, does this mean there is an issue with disk I/O?
>>
>> strace: Process 5134 attached
>>
>> epoll_wait(3, [{EPOLLIN, {u32=31860224, u64=31860224}}], 1, -1) = 1
>>
>> close(3) = 0
>>
>> recvfrom(10, "\2\0\0\0\230\0\0\0\7@\0\0\1\0\0\0\5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
>> 1000, 0, NULL, NULL) = 152
>>
>> recvfrom(10, 0x7ffeeb967fa0, 1000, 0, NULL, NULL) = -1 EAGAIN (Resource
>> temporarily unavailable)
>>
>> epoll_create1(EPOLL_CLOEXEC) = 3
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 11, {EPOLLIN|EPOLLERR|EPOLLHUP,
>> {u32=31860176, u64=31860176}}) = 0
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 7, {EPOLLIN|EPOLLERR|EPOLLHUP, {u32=31860200,
>> u64=31860200}}) = 0
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 10, {EPOLLIN|EPOLLERR|EPOLLHUP,
>> {u32=31860224, u64=31860224}}) = 0
>>
>> epoll_wait(3, [{EPOLLIN, {u32=31860224, u64=31860224}}], 1, -1) = 1
>>
>> close(3) = 0
>>
>> recvfrom(10, "\2\0\0\0\250\3\0\0\7@\0\0\10\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
>> 1000, 0, NULL, NULL) = 936
>>
>> recvfrom(10,
>> "\2\0\0\0\250\3\0\0\0\0\0\0\10\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
>> 1000, 0, NULL, NULL) = 936
>>
>> recvfrom(10, "\2\0\0\0x\1\0\0\7@\0\0\3\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
>> 1000, 0, NULL, NULL) = 376
>>
>> recvfrom(10, 0x7ffeeb967fa0, 1000, 0, NULL, NULL) = -1 EAGAIN (Resource
>> temporarily unavailable)
>>
>> epoll_create1(EPOLL_CLOEXEC) = 3
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 11, {EPOLLIN|EPOLLERR|EPOLLHUP,
>> {u32=31860176, u64=31860176}}) = 0
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 7, {EPOLLIN|EPOLLERR|EPOLLHUP, {u32=31860200,
>> u64=31860200}}) = 0
>>
>> epoll_ctl(3, EPOLL_CTL_ADD, 10, {EPOLLIN|EPOLLERR|EPOLLHUP,
>> {u32=31860224, u64=31860224}}) = 0
>>
>> epoll_wait(3, [{EPOLLIN, {u32=31860224, u64=31860224}}], 1, -1) = 1
>>
>> close(3)
>>
>>
>> Regards,
>>
>> Nikhil
>>
>> On Wed, Apr 27, 2022 at 8:25 PM Adelino Silva <[email protected]>
>> wrote:
>>
>> One possible cause for this problem is that the statistics collector
>> process is bound to an IP:port which is not responding.
>> See the following thread discussion.
>>
>>
>> https://stackoverflow.com/questions/46008372/using-stale-statistics-instead-of-current-ones
>>
>> <https://stackoverflow.com/questions/46008372/using-stale-statistics-instead-of-current-ones;
>> Using stale statistics instead of current ones - Stack Overflow
>> <https://stackoverflow.com/questions/46008372/using-stale-statistics-instead-of-current-ones;
>> Teams. Q&A for work. Connect and share knowledge within a single location
>> that is structured and easy to search. Learn more
>> stackoverflow.com
>>
>> Regards,
>>
>> Adelino Silva
>>
>> ------------------------------
>> *From:* Nikhil Shetty <[email protected]>
>> *Sent:* Wednesday, April 27, 2022 2:49 PM
>> *To:* Adelino Silva <[email protected]>
>> *Cc:* Pgsql-admin <[email protected]>
>> *Subject:* [EXTERNAL] Re: Postgres Stale Statistics
>>
>> Hi Adelino, I had gone through that thread before, we cannot move the
>> stats to RAM as of now. Thanks, Nikhil On Wed, Apr 27, 2022 at 6:16 PM
>> Adelino Silva <[email protected]> wrote: Hi, Found this thread
>> that explains the warning.
>> ZjQcmQRYFpfptBannerStart
>> This Message Is From an External Sender
>> This message came from outside your organization.
>>
>> ZjQcmQRYFpfptBannerEnd
>> Hi Adelino,
>>
>> I had gone through that thread before, we cannot move the stats to RAM as
>> of now.
>>
>> Thanks,
>> Nikhil
>>
>> On Wed, Apr 27, 2022 at 6:16 PM Adelino Silva <[email protected]>
>> wrote:
>>
>> Hi,
>>
>> Found this thread that explains the warning.
>> using stale statistics instead of current ones because stats collector is
>> not responding
>>
>> https://www.postgresql.org/message-id/[email protected]
>>
>> <https://www.postgresql.org/message-id/[email protected];
>> PostgreSQL: Re: using stale statistics instead of current ones because
>> stats collector is not responding
>> <https://www.postgresql.org/message-id/[email protected];
>> Hi, On Tue, 2016-03-08 at 16:18 -0800, Tory M Blue wrote: > No hits on
>> the intratubes on this. > …
>> www.postgresql.org
>>
>>
>> Regards,
>>
>> Adelino Silva
>>
>> ------------------------------
>> *From:* Nikhil Shetty <[email protected]>
>> *Sent:* Wednesday, April 27, 2022 12:08 PM
>> *To:* Pgsql-admin <[email protected]>
>> *Subject:* [EXTERNAL] Postgres Stale Statistics
>>
>> Hi, We are getting below WARNING on one of the standby instances. Not
>> sure what caused it but to resolve it we tried restarting the database
>> instances but it is still not working WARNING - using stale statistics
>> instead of current ones because
>> ZjQcmQRYFpfptBannerStart
>> This Message Is From an External Sender
>> This message came from outside your organization.
>>
>> ZjQcmQRYFpfptBannerEnd
>> Hi,
>>
>> We are getting below WARNING on one of the standby instances. Not sure
>> what caused it but to resolve it we tried restarting the database instances
>> but it is still not working
>>
>>
>> WARNING - using stale statistics instead of current ones because stats
>> collector is not responding
>>
>>
>> Postgresql version - 11.7
>>
>>
>> Any other option to resolve this? We are thinking of building the standby
>> again but what if the WARNING is for a primary database instance and a
>> restart won't solve it?
>>
>>
>> Thanks and Regards,
>>
>> Nikhil
>>
>>
view thread (9+ messages)
reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: [email protected]
Cc: [email protected], [email protected], [email protected]
Subject: Re: Postgres Stale Statistics
In-Reply-To: <CAFpL5VyuYPS9on8ANA8vQ4Fdr18NWGvdGcD2YnB-ZdVFtPHwvg@mail.gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox