Обсуждение: doc fix for pg_stat_activity.backend_type
Hi all, Commit fc70a4b0df3 added backend_type to pg_stat_activity, but the documentation omitted "logical replication launcher". Patch attached. -John Naylor
Вложения
On Mon, Nov 12, 2018 at 1:20 PM John Naylor <jcnaylor@gmail.com> wrote: > > Hi all, > > Commit fc70a4b0df3 added backend_type to pg_stat_activity, but the > documentation omitted "logical replication launcher". Patch attached. > Isn't this the fallout of commit 5373bc2a08 which has added background worker types? If so, I guess your patch needs modification. -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
On 11/12/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > On Mon, Nov 12, 2018 at 1:20 PM John Naylor <jcnaylor@gmail.com> wrote: >> >> Hi all, >> >> Commit fc70a4b0df3 added backend_type to pg_stat_activity, but the >> documentation omitted "logical replication launcher". Patch attached. >> > > Isn't this the fallout of commit 5373bc2a08 which has added background > worker types? If so, I guess your patch needs modification. Looks like it. A quick search revealed "parallel worker" and "logical replication worker". src/test/modules/ also show "test_shm_mq" and "worker_spi", but it seems those don't need to be publicly documented. If that sounds right I'll update the patch to include the first two. -John Naylor
On Mon, Nov 12, 2018 at 09:42:45PM +0700, John Naylor wrote: > Looks like it. A quick search revealed "parallel worker" and "logical > replication worker". src/test/modules/ also show "test_shm_mq" and > "worker_spi", but it seems those don't need to be publicly documented. > If that sounds right I'll update the patch to include the first two. Just wondering: do we actually need to include in the docs this list at all? This is a recipe to forget its update each time a new backend type is added. -- Michael
Вложения
On Tue, Nov 13, 2018 at 5:38 AM Michael Paquier <michael@paquier.xyz> wrote: > > On Mon, Nov 12, 2018 at 09:42:45PM +0700, John Naylor wrote: > > Looks like it. A quick search revealed "parallel worker" and "logical > > replication worker". src/test/modules/ also show "test_shm_mq" and > > "worker_spi", but it seems those don't need to be publicly documented. > > If that sounds right I'll update the patch to include the first two. > > Just wondering: do we actually need to include in the docs this list at > all? This is a recipe to forget its update each time a new backend type > is added. > Sure, but how will we justify documenting (autovacuum launcher and autovacuum worker) and not (logical replication launcher and logical replication worker)? I think we can document the type of workers that are part of core-server functionality. We can make some generic statement on the workers that can be launched by extensions. -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > On Tue, Nov 13, 2018 at 5:38 AM Michael Paquier <michael@paquier.xyz> > wrote: >> >> On Mon, Nov 12, 2018 at 09:42:45PM +0700, John Naylor wrote: >> > Looks like it. A quick search revealed "parallel worker" and "logical >> > replication worker". src/test/modules/ also show "test_shm_mq" and >> > "worker_spi", but it seems those don't need to be publicly documented. >> > If that sounds right I'll update the patch to include the first two. >> >> Just wondering: do we actually need to include in the docs this list at >> all? This is a recipe to forget its update each time a new backend type >> is added. >> > > Sure, but how will we justify documenting (autovacuum launcher and > autovacuum worker) and not (logical replication launcher and logical > replication worker)? I think we can document the type of workers that > are part of core-server functionality. We can make some generic > statement on the workers that can be launched by extensions. How about something like the attached? -John Naylor
Вложения
On Tue, Nov 13, 2018 at 12:04 PM John Naylor <jcnaylor@gmail.com> wrote: > > On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > > On Tue, Nov 13, 2018 at 5:38 AM Michael Paquier <michael@paquier.xyz> > > wrote: > >> > >> On Mon, Nov 12, 2018 at 09:42:45PM +0700, John Naylor wrote: > >> > Looks like it. A quick search revealed "parallel worker" and "logical > >> > replication worker". src/test/modules/ also show "test_shm_mq" and > >> > "worker_spi", but it seems those don't need to be publicly documented. > >> > If that sounds right I'll update the patch to include the first two. > >> > >> Just wondering: do we actually need to include in the docs this list at > >> all? This is a recipe to forget its update each time a new backend type > >> is added. > >> > > > > Sure, but how will we justify documenting (autovacuum launcher and > > autovacuum worker) and not (logical replication launcher and logical > > replication worker)? I think we can document the type of workers that > > are part of core-server functionality. We can make some generic > > statement on the workers that can be launched by extensions. > > How about something like the attached? > Don't you need to remove <literal>background worker</literal>? + In addition, extensions may have additional types. How about: "In addition, background workers registered by extensions may have additional types."? -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > On Tue, Nov 13, 2018 at 12:04 PM John Naylor <jcnaylor@gmail.com> wrote: >> >> On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: >> > On Tue, Nov 13, 2018 at 5:38 AM Michael Paquier <michael@paquier.xyz> >> > wrote: >> >> >> >> On Mon, Nov 12, 2018 at 09:42:45PM +0700, John Naylor wrote: >> >> > Looks like it. A quick search revealed "parallel worker" and >> >> > "logical >> >> > replication worker". src/test/modules/ also show "test_shm_mq" and >> >> > "worker_spi", but it seems those don't need to be publicly >> >> > documented. >> >> > If that sounds right I'll update the patch to include the first two. >> >> >> >> Just wondering: do we actually need to include in the docs this list >> >> at >> >> all? This is a recipe to forget its update each time a new backend >> >> type >> >> is added. >> >> >> > >> > Sure, but how will we justify documenting (autovacuum launcher and >> > autovacuum worker) and not (logical replication launcher and logical >> > replication worker)? I think we can document the type of workers that >> > are part of core-server functionality. We can make some generic >> > statement on the workers that can be launched by extensions. >> >> How about something like the attached? >> > > Don't you need to remove <literal>background worker</literal>? It's handled in pgstat_get_backend_desc(), so I assumed not. If that's just a place holder, then it's probably better left out, as in the attached. > + In addition, extensions may have additional types. > > How about: "In addition, background workers registered by extensions > may have additional types."? Sounds good to me -- I've included this language. -John Naylor
Вложения
On Tue, Nov 13, 2018 at 3:37 PM John Naylor <jcnaylor@gmail.com> wrote: > > On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > > > > Don't you need to remove <literal>background worker</literal>? > > It's handled in pgstat_get_backend_desc(), so I assumed not. If that's > just a place holder, then it's probably better left out, as in the > attached. > I don't think 'background worker' can be displayed as backend_type. Do you see any way it can be displayed as backend_type? > > + In addition, extensions may have additional types. > > > > How about: "In addition, background workers registered by extensions > > may have additional types."? > > Sounds good to me -- I've included this language. > LGTM. I will wait for a day or so, if nobody has any comments, I will push your patch. -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > On Tue, Nov 13, 2018 at 3:37 PM John Naylor <jcnaylor@gmail.com> wrote: >> >> On 11/13/18, Amit Kapila <amit.kapila16@gmail.com> wrote: >> > >> > Don't you need to remove <literal>background worker</literal>? >> >> It's handled in pgstat_get_backend_desc(), so I assumed not. If that's >> just a place holder, then it's probably better left out, as in the >> attached. >> > > I don't think 'background worker' can be displayed as backend_type. I think you're right (pgstatfuncs.c, starting at line 826). -John Naylor
On Tue, Nov 13, 2018 at 5:03 PM Amit Kapila <amit.kapila16@gmail.com> wrote: > > On Tue, Nov 13, 2018 at 3:37 PM John Naylor <jcnaylor@gmail.com> wrote: > > > + In addition, extensions may have additional types. > > > > > > How about: "In addition, background workers registered by extensions > > > may have additional types."? > > > > Sounds good to me -- I've included this language. > > > > LGTM. I will wait for a day or so, if nobody has any comments, I will > push your patch. > Pushed, thanks for working on this. -- With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
On 11/15/18, Amit Kapila <amit.kapila16@gmail.com> wrote: > Pushed, thanks for working on this. Thanks for your input. -John Naylor