mitted v6 instead. Let's see
> what breaks...
>
There was a duplicate error check for the invalid createdb strategy
option in the test case, although it would not create any issue but it
is duplicate so I have fixed it in the attached patch.
--
Regards,
Dilip Kumar
EnterpriseDB: http://ww
ngle test for it.
I will add tests for both of these cases and send the patch.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Thu, Mar 31, 2022 at 9:46 AM Dilip Kumar wrote:
>
> On Thu, Mar 31, 2022 at 5:07 AM Andres Freund wrote:
> >
> > Hi,
> >
> > On 2022-03-29 11:55:05 -0400, Robert Haas wrote:
> > > I committed v6 instead.
> >
> > I was just discussing t
vacuuming certain partitions when we go for global index vacuuming we
don't want to rescan all the partitions to get the same dead items) b)
and even without global indexes there are advantages of storing dead
items in the conveyor belt as explained in my previous paragraph. So
I think it is worth
d verify it then they can do so. But with the current
tests in mind the way I got it in the attached patch has less
duplicate code so I preferred it this way.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
From a672a27a9e502331a7c4ca2a16f3f660c8ed3fcd Mon Sep 17 00:00:00 2001
F
there we don't have a memory context to
> reset afaics. Oddly enough, it sure looks like we have an existing version of
> this bug in the file-copy path?
Yeah, I see that the createdb() and dbase_redo() had this existing
problem and with this patch we have created a few more such
oc
so that we do not create an unnecessary
directory under the tablespace where we do not have any data to be
copied.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Fri, Apr 1, 2022 at 11:34 PM Robert Haas wrote:
>
> On Fri, Apr 1, 2022 at 12:08 AM Dilip Kumar wrote:
> > After thinking more about this I see there is some value of
> > remembering the dead tids in the conveyor belt. Basically, the point
> > is if there are multip
ser has to create and what will be replicated using
DDL replication? Suppose the user takes a dump and copies all the
schema and then creates the subscription, then how we are we going to
handle the DDL concurrent to the subscription command?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Fri, Apr 8, 2022 at 4:30 PM Amit Kapila wrote:
>
> On Tue, Mar 29, 2022 at 9:47 AM Dilip Kumar wrote:
> > >
> > > The idea is to force skipping any direct data population (which can
> > > potentially cause data inconsistency on the subscriber)
> > > i
be easily reproducible every time. If this happens to be true then
> I think your suggestion related to increasing autovacuum_naptime would
> work.
>
I will try to reproduce this, maybe by reducing the autovacuum_naptime or
parallelly running some script that continuously performs DDL-only
transactions.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Apr 18, 2022 at 3:29 PM Dilip Kumar wrote:
>
> This needs to be verified once by doing some manual testing as it may
> not be easily reproducible every time. If this happens to be true then
> I think your suggestion related to increasing autovacuum_naptime would
> work.
&
m under the same buffer
pool and comparing against relation pages for victim buffer selection
might cause different problems. But anyway I would discuss those
points maybe in that thread.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Tue, Oct 24, 2023 at 9:34 PM Alvaro Herrera wrote:
>
> On 2023-Oct-11, Dilip Kumar wrote:
>
> > In my last email, I forgot to give the link from where I have taken
> > the base path for dividing the buffer pool in banks so giving the same
> > here[1]. And looking
On Wed, Oct 25, 2023 at 5:58 PM Amit Kapila wrote:
>
> On Fri, Oct 20, 2023 at 9:40 AM Dilip Kumar wrote:
> >
> > On Sat, Oct 14, 2023 at 9:43 AM Amit Kapila wrote:
> > >
> > > This and other results shared by you look promising. Will there be any
> >
pages
for the actual backend and the worker? I mean here for the backend you
are showing lossy pages only if it is > 0 whereas for workers we are
showing 0 lossy pages as well.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Sun, Nov 5, 2023 at 1:37 AM Andrey M. Borodin wrote:
> On 30 Oct 2023, at 09:20, Dilip Kumar wrote:
>
> changed the logic of SlruAdjustNSlots() in 0002, such that now it
> starts with the next power of 2 value of the configured slots and
> keeps doubling the number of banks
On Mon, Nov 6, 2023 at 1:05 PM Andrey M. Borodin wrote:
> > On 6 Nov 2023, at 09:09, Dilip Kumar wrote:
> >
> >
> >> Having hashtable to find SLRU page in the buffer IMV is too slow. Some
> >> comments on this approach can be found here [0].
> >> I&
k partitioning. Simplehash has not, AFAIK.
Yeah, Simplehash doesn't have partitioning so with simple hash we will
be stuck with the centralized control lock that is one of the main
problems trying to solve here.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
rom this. Although I don't have any number which proves this.
Currently, I want to focus on all the base patches and keep this patch
as add on and later if we find its useful and want to pursue this then
we will see how to make it better readable.
>
> Subject: [PATCH v4 5/5] Ensure slru buffer slots are in multiple of numbe of
> partitions
>
> I think the 0005 patch can be merged to 0001.
Yeah in the next version, it is done that way. Planning to post end of the day.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Thu, Nov 9, 2023 at 9:39 PM Robert Haas wrote:
>
> On Wed, Nov 8, 2023 at 6:41 AM Dilip Kumar wrote:
> > Here is the updated version of the patch, here I have taken the
> > approach suggested by Andrey and I discussed the same with Alvaro
> > offlist and he also agre
; them alone and only adjust the one with which they have a performance
> problems; it's not going to be the same one for everybody.
+1
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
es < 0)
+ ereport(ERROR,
+ (errcode_for_file_access(),
+ errmsg("could not write file \"%s\": %m",
+ FilePathName(io->file;
/could not write file/ could not read file
8.
+/*
+ * Comparator to sort a List of WalSummaryFile objects by start_lsn.
+ */
+static int
+ListComparatorForWalSummaryFiles(const ListCell *a, const ListCell *b)
+{
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Tue, Nov 14, 2023 at 12:52 AM Robert Haas wrote:
>
> On Fri, Nov 10, 2023 at 6:27 AM Dilip Kumar wrote:
> > - I think 0001 looks good improvement irrespective of the patch series.
>
> OK, perhaps that can be independently committed, then, if nobody objects.
>
> Tha
already returning 'BlockRefTableEntry' then why
would it need to set an extra '*limit_block' out parameter which it is
actually reading from the entry itself?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
> 64k buffers, since useful values are below a couple thousand anyhow.
I agree, that allowing xact_buffers to grow beyond 65536 up to the
slru.h-defined limit of 131072 is not that bad, so I will change that
in the next version.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
so I'm doubly hesitant to go
> without any testing here. Maybe it'd be possible to use Michael
> Paquier's injection points somehow?
Sorry, but I am not aware of "Michael Paquier's injection" Is it
something already in the repo? Can you redirect me to some of
ther
external functions starting with "SimpleLruGet", are you fine with
this name?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Fri, Nov 17, 2023 at 7:28 PM Alvaro Herrera wrote:
>
> On 2023-Nov-17, Dilip Kumar wrote:
I think I need some more clarification for some of the review comments
> > On Thu, Nov 16, 2023 at 3:11 PM Alvaro Herrera
> > wrote:
> > >
> > > I just no
On Mon, Nov 20, 2023 at 2:37 PM Andrey M. Borodin wrote:
> > On 20 Nov 2023, at 13:51, Dilip Kumar wrote:
> >
> > 2) Do we really need one separate lwlock tranche for each SLRU?
> >
> > IMHO if we use the same lwlock tranche then the wait event will show
> >
On Tue, Nov 21, 2023 at 2:03 PM Dilip Kumar wrote:
>
> On Mon, Nov 20, 2023 at 4:42 PM Dilip Kumar wrote:
> >
> > On Mon, Nov 20, 2023 at 2:37 PM Andrey M. Borodin
> > wrote:
> >
> > > > On 20 Nov 2023, at 13:51, Dilip Kumar wrote:
> > >
On Thu, Nov 23, 2023 at 11:34 AM Dilip Kumar wrote:
>
> Note: With this testing, we have found a bug in the bank-wise
> approach, basically we are clearing a procglobal->clogGroupFirst, even
> before acquiring the bank lock that means in most of the cases there
> will be a sing
async.h:15
> > error: src/include/commands/async.h: patch does not apply
>
> Yeah, this patch series conflicts with today's commit 4ed8f0913bfd.
I will send a rebased version by tomorrow.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
he queue of the non-transactional sequence change so
that while sending it to downstream whenever it is necessary we will
change the historic snapshot?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Wed, Dec 6, 2023 at 11:12 AM Dilip Kumar wrote:
>
> On Sun, Dec 3, 2023 at 11:22 PM Tomas Vondra
> wrote:
> >
I was also wondering what happens if the sequence changes are
transactional but somehow the snap builder state changes to
SNAPBUILD_FULL_SNAPSHOT in between pr
al sequence changes in a separate queue in
the order they arrive and as soon as we process the next commit we
will process all the non-transactional changes at that time. Do you
see issue with that?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Wed, Dec 6, 2023 at 7:17 PM Tomas Vondra
wrote:
>
> On 12/6/23 12:05, Dilip Kumar wrote:
> > On Wed, Dec 6, 2023 at 3:36 PM Amit Kapila wrote:
> >>
> >>> Why can't we use the same concept of
> >>> SnapBuildDistributeNewCatalogSnapshot(),
bility of the
upcoming operation in the same transaction, but that's not an issue
because if some of the changes of this transaction are seen when
snapbuild state < SNAPBUILD_FULL_SNAPSHOT then this transaction has to
get committed before the state change to SNAPBUILD_CONSISTENT_SNAPSHOT
i.
b4_%3DzNoKU%2B8FP-S6zpv-r4Gm-Y%2BQ%40mail.gmail.com
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
ould not initiate base backup: ERROR: manifest
requires WAL from final timeline 1 ending at 0/6F8, but this
backup starts at 0/628
pg_basebackup: removing data directory "d1"
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Dec 11, 2023 at 11:44 AM Dilip Kumar wrote:
>
> On Tue, Dec 5, 2023 at 11:40 PM Robert Haas wrote:
> >
> > On Mon, Dec 4, 2023 at 3:58 PM Robert Haas wrote:
> > > Considering all this, what I'm inclined to do is go and put
> > > UPLOAD_MANIFE
ould be the user's responsibility. I mean if the user has set
'standby_slot_names' on standby then let standby also wait for
cascading standby to sync their slots? Is there any issue with that
behavior?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Dec 11, 2023 at 2:21 PM shveta malik wrote:
>
> On Mon, Dec 11, 2023 at 1:47 PM Dilip Kumar wrote:
> >
> > On Fri, Dec 8, 2023 at 2:36 PM Amit Kapila wrote:
> > >
> > > On Wed, Dec 6, 2023 at 4:53 PM shveta malik
> > > wrote:
> > >
if we frequently select
many of these transactions to come under memory limit instead of
selecting a couple of large transactions which are consuming 8-9%?
>
> As a side note, the idea (c) mentioned in the comment, evicting
> multiple transactions at once to free a given portion of the memory,
> would also help in avoiding back and forth the memory threshold. It's
> also worth considering.
Yes, I think it is worth considering.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Dec 11, 2023 at 3:14 PM Michael Paquier wrote:
>
> On Mon, Dec 11, 2023 at 11:09:45AM +0530, Dilip Kumar wrote:
> > I haven't specifically done a review or testing of this patch, but I
> > have used this for testing the CLOG group update code with my
> > SLRU
xceptional case we are checking the respective bank
lock for each page and if that exception occurred we will release the
old bank lock and acquire a new lock. This case might not be
performant because now it is possible that after getting the lock
leader might need to wait again on another bank
ween the
largest and smallest transaction in this list are not very
significant. And remaining in the small transaction list and from the
small transaction list we can choose to spill multiple transactions at
a time.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
ate does
not reach FULL just add the sequence relids to the hash, I mean that
hash is only maintained for deciding whether the sequence is changed
in that transaction or not. So no adding such relids to hash seems
like a root cause of the issue. Honestly, I haven't analyzed this
idea in det
On Thu, Dec 14, 2023 at 12:31 PM Ashutosh Bapat
wrote:
>
> On Thu, Dec 14, 2023 at 10:53 AM Dilip Kumar wrote:
>
> > > >
> > >
> > > It is correct that we can make a wrong decision about whether a change
> > > is transactional or non-transact
4| 512/1024| 44
patch-2 | 1024| 512/1024| 72
[1]
https://www.postgresql.org/message-id/flat/e46cdea96979545b2d8a13b451d8b1ce61dc7238.camel%40postgrespro.ru#0ed2cad11470825d464093fe6b8ef6a3
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Thu, Dec 14, 2023 at 4:36 PM Dilip Kumar wrote:
>
> On Wed, Dec 13, 2023 at 5:49 PM Andrey M. Borodin
> wrote:
>
> > > On 12 Dec 2023, at 18:28, Alvaro Herrera wrote:
> > >
> > > Andrey, do you have any stress tests or anything else that you use
client counts. So as per
my understanding by distributing the SLRU locks there are scenarios
where we will not need group update anymore but OTOH there are also
scenarios where we will still benefit from the group update.
[1]
https://www.postgresql.org/message-id/CAFiTN-u-XEzhd%3DhNGW586f
forward direction?
+ */
+ for (int i = 1; i < PREFETCH_SEQ_PATTERN_BLOCKS; i++)
Correct, I think if we fetch this forward it will have an advantage
with memory prefetching.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
want to rethink
> the way the SLRU is extended -- page at a time wouldn't really make
> sense, but preallocating an entire file might.
Yeah, this is indeed an interesting idea. So I think if we are
interested in working in this direction maybe this can be submitted in
a different thread, IMHO.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
usage then
we can expand the behavior by providing an additional GUC with dbname.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Tue, Dec 12, 2023 at 4:15 PM Michael Paquier wrote:
>
> On Tue, Dec 12, 2023 at 10:27:09AM +0530, Dilip Kumar wrote:
> > Oops, I only included the code changes where I am adding injection
> > points and some comments to verify that, but missed the actual test
> > file.
ING, and ENABLED. The state transition process
+ * between these values is the same as the two_phase option (see TWO_PHASE
+ * TRANSACTIONS for details).
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Fri, Jan 5, 2024 at 5:45 PM Amit Kapila wrote:
>
> On Fri, Jan 5, 2024 at 4:25 PM Dilip Kumar wrote:
> >
> > On Fri, Jan 5, 2024 at 8:59 AM shveta malik wrote:
> > >
> > I was going the the patch set again, I have a question. The below
> > comments
nk there is a problem with multiple processes getting the
lock? I mean they are modifying different CLOG pages so that can be
done concurrently right?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
replication slot, which allows informing the primary about the xmin and
+ * catalog_xmin values on the standby.
I do not like capitalizing the first letter of the
'hot_standby_feedback' which is a GUC parameter
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Wed, Aug 30, 2023 at 1:03 PM Michael Paquier wrote:
>
> On Mon, Aug 28, 2023 at 01:47:18PM +0530, Dilip Kumar wrote:
> > I removed this mainly because now in other comments[1] where we are
> > introducing this new CHECKPOINT_REDO record we are explaining the
> > p
On Thu, Aug 31, 2023 at 9:36 AM Michael Paquier wrote:
>
> On Wed, Aug 30, 2023 at 04:51:19PM +0530, Dilip Kumar wrote:
> > Your suggestions LGTM so modified accordingly
>
> I have been putting my HEAD on this patch for a few hours, reviewing
> the surroundings, and some
,
> >> but I plan to commit both patches sometime today.
> >
> > Thanks! I'm glad your committing the patches, and I approve of you
> > spending way too much time on the commit message. :-)
>
> Committed.
Sorry, I didn't notice this thread earlier. The new
the
slots. If it is printing the name of the slots then shouldn't it be
printing all the slots' names or it should just say that there
existing slots on the new cluster without giving any names? And if we
are planning for option 2 i.e. not printing the name then better to
put LIMIT 1 at the end of the query.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
ting the name of the slots then shouldn't it be
> > printing all the slots' names or it should just say that there
> > existing slots on the new cluster without giving any names? And if we
> > are planning for option 2 i.e. not printing the name then better to
> > put LIMIT 1 at the end of the query.
>
> I'm planning to change that the number of slots are reported by using
> count(*).
Yeah, that seems a better option.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Fri, Sep 1, 2023 at 9:47 AM Dilip Kumar wrote:
>
> On Thu, Aug 31, 2023 at 7:56 PM Hayato Kuroda (Fujitsu)
> wrote:
>
Some more comments on 0002
1.
+ conn = connectToServer(&new_cluster, "template1");
+
+ prep_status("Checking for logical replication slots"
very rare that
some of the segments are getting dropped just during the upgrade time
and that too from the old cluster so in such cases not migrating the
slots which are invalidated should be fine no?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
e: "pg_catalog.pg_largeobject", reloid: 2613, reltblspace: ""
> relname: "pg_catalog.pg_largeobject_loid_pn_index", reloid: 2683,
> reltblspace: ""
> Logical replication slots within the database:
> slotname: "old1", plugin: "test_decoding", two_phase: 0
> slotname: "old2", plugin: "test_decoding", two_phase: 0
> slotname: "old3", plugin: "test_decoding", two_phase: 0
Yeah this looks good now.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Sep 4, 2023 at 8:41 AM Amit Kapila wrote:
>
> On Sat, Sep 2, 2023 at 6:12 PM Dilip Kumar wrote:
> >
> > On Sat, Sep 2, 2023 at 10:09 AM Amit Kapila wrote:
> >
> > > The other possibilities apart from not allowing an upgrade in such a
> > >
On Mon, Sep 4, 2023 at 11:18 AM Amit Kapila wrote:
>
> On Mon, Sep 4, 2023 at 10:33 AM Dilip Kumar wrote:
> >
> > On Mon, Sep 4, 2023 at 8:41 AM Amit Kapila wrote:
> > >
> > > On Sat, Sep 2, 2023 at 6:12 PM Dilip Kumar wrote:
> > > >
>
lararraysel(PlannerInfo *root,
ObjectIdGetDatum(operator),
PointerGetDatum(args),
Int32GetDatum(varRelid)));
-
+ list_free(args);
+ pfree(c);
Maybe you can just use list_free_deep, instead of storing the constant
in a separate variable.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Sep 4, 2023 at 1:41 PM Dilip Kumar wrote:
>
> > > I think we can do better, like we can just read the latest
> > > checkpoint's LSN before starting the old cluster. And now while
> > > checking the slot can't we check if the the slot is invalida
args),
> > Int32GetDatum(varRelid)));
> > -
> > + list_free(args);
> > + pfree(c);
> >
> > Maybe you can just use list_free_deep, instead of storing the constant
> > in a separate variable.
> As I see, the first element in the array is leftop, whic
y accessible, do they have the right contents?
Can't we think of comparing at the block level, like we can compare
each block but ignore the content of the hole?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
n the thread and we have a
conclusion on this then maybe we can mention this in comments?
/*
* Flush all replication slots to disk.
*
- * This needn't actually be part of a checkpoint, but it's a convenient
- * location.
+ * is_shutdown is true in case of a shutdown checkpoint.
*/
void
-CheckPointReplicationSlots(void)
+CheckPointReplicationSlots(bool is_shutdown)
It seems we have removed two lines from the function header comments,
is this intentional or accidental?
Other than this patch LGTM.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Tue, Sep 5, 2023 at 9:38 AM Amit Kapila wrote:
>
> On Mon, Sep 4, 2023 at 4:19 PM Dilip Kumar wrote:
> >
> > Said that there is a possibility that some of the slots which got
> > invalidated even on the previous checkpoint might get the same LSN as
> > the slot
On Tue, Sep 5, 2023 at 9:58 AM Amit Kapila wrote:
>
> On Tue, Sep 5, 2023 at 9:10 AM Dilip Kumar wrote:
> >
> > The comments don't mention anything about why we are just flushing at
> > the shutdown checkpoint. I mean the checkpoint is not that frequent
> >
ng normal operation?
Okay, I am also not sure of that.
If users really want to leave
> behind such invalidated slots after upgrade, we can even think of
> providing some option like "exclude_invalid_logical_slots".
+1
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
k and call
> > > ReplicationSlotMarkDirty which will take lock again.
> >
> > Yes, this is unavoidable, but maybe it's not a big problem as
> > we only do it at shutdown.
> >
>
> True but still it doesn't look elegant. I also thought about
lock would
> make
> it more robust.
I think our all logic is based on that the walsender is existed
already. If not then even if you check under the mutex that the
confirmed flush LSN is not changed then it can changed right after you
release the lock and then we will not be flushing the lat
On Wed, Sep 6, 2023 at 9:47 AM Amit Kapila wrote:
>
> On Tue, Sep 5, 2023 at 6:00 PM Dilip Kumar wrote:
>
> Right, it can change and in that case, the check related to
> confirm_flush LSN will fail during the upgrade. However, the point is
> that if we don't take spinlo
On Wed, Sep 6, 2023 at 12:01 PM Amit Kapila wrote:
>
> On Wed, Sep 6, 2023 at 9:57 AM Dilip Kumar wrote:
> >
> > On Wed, Sep 6, 2023 at 9:47 AM Amit Kapila wrote:
> > >
> > > On Tue, Sep 5, 2023 at 6:00 PM Dilip Kumar wrote:
> >
> > >
>
o be more
> > defensive. But couldn't we do that just like we do for autovacuum and
> > force the GUCs that could remove the slot's WAL to not do their work
> > instead?
> >
>
> I think if we just make max_slot_wal_keep_size to -1 that should be
> sufficient to not let any slots get invalidated during upgrade. Do you
> have anything else in mind?
This seems like a good solution to the problem.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
here, especially the words
RECHECK, DISABLE, and ENABLE. And also the name of the enum is a bit
off. We can name it more like a unique constraint
YES, PARTIAL, EXISTING
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
isBinaryUpgrade is tempting in the invalidation slot path.
> >
>
> I agree with doing something simple. So, to conclude, we agree on two
> things in this thread (a) Use max_slot_wal_keep_size to -1 to start
> postmaster for the old cluster during the upgrade; (b) Have an
> elog(ERROR) to avoid invalidating slots during the upgrade.
+1
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
he slot still usable?" I think it should be "Is
the slot usable?" otherwise it appears that we have first fetched the
slots and now we are refetching it and checking whether it is still
usable.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Sep 11, 2023 at 11:16 AM Amit Kapila wrote:
>
> On Mon, Sep 11, 2023 at 10:39 AM Dilip Kumar wrote:
> >
> > 3.
> > - with the primary.) Replication slots are not copied and must
> > - be recreated.
> > + with the primary.) Replicat
remental_blocks)
+ break;
6.
+typedef struct
+{
+ TimeLineID tli;
+ XLogRecPtr start_lsn;
+ XLogRecPtr end_lsn;
+} backup_wal_range;
+
+typedef struct
+{
+ uint32 status;
+ const char *path;
+ size_t size;
+} backup_file_entry;
Better we add some comments for these structures.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
and again shut it down? I don't think
that is really an acceptable idea. Maybe as discussed in the past we
can provide an option to skip the slot checking and during the --check
command we can give a WARNING and suggest that better to use
--skip-slot-checking for the main upgrade as we have already checked.
This could still be okay for the user.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
g_upgrade, might need some agreement.
Yeah right, in fact during the --check command we can give that
suggestion as well.
I feel option 2 looks best to me unless there is some design issue to
that, as of now I do not see any issue with that though. Let's see
what others think.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Thu, Sep 14, 2023 at 10:00 AM Amit Kapila wrote:
>
> On Thu, Sep 14, 2023 at 9:21 AM Dilip Kumar wrote:
> > > Cons: Although we have some examples for using functions
> > > (binary_upgrade_set_next_pg_enum_oid ...) to set some variables during
> > > upgra
2_VACUUM or other vacuum-related commands
can not occur during the upgrade?
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Wed, Sep 20, 2023 at 12:12 PM Amit Kapila wrote:
>
> On Wed, Sep 20, 2023 at 11:51 AM Dilip Kumar wrote:
> >
> > On Wed, Sep 20, 2023 at 11:00 AM Hayato Kuroda (Fujitsu)
> > wrote:
> > >
> > > Dear Amit,
> > >
> > > Thank you fo
origptr);
}
}
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
if (XLOG_CHECKPOINT_REDO) should be else if (info == XLOG_CHECKPOINT_REDO)
That's all I have for now.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Wed, Sep 20, 2023 at 3:23 PM Dilip Kumar wrote:
>
> On Wed, Aug 16, 2023 at 7:57 PM Tomas Vondra
> wrote:
> >
>
> I was reading through 0001, I noticed this comment in
> ReorderBufferSequenceIsTransactional() function
>
> + * To decide if a sequence change shou
since the last checkpoint and that is not yet synched to the disk then
we are doing so. I think this is the most important change otherwise
many slots for which we have already streamed all the WAL might give
an error assuming that there are pending WAL from the slots which are
not yet confirmed.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
On Mon, Sep 25, 2023 at 12:30 PM Dilip Kumar wrote:
>
> On Mon, Sep 25, 2023 at 11:15 AM Bharath Rupireddy
> wrote:
> >
> > On Fri, Sep 22, 2023 at 12:11 PM Amit Kapila
> > wrote:
> > >
> > > Yeah, both by tests and manually verifying the WAL re
On Mon, Sep 25, 2023 at 1:23 PM Bharath Rupireddy
wrote:
>
> On Mon, Sep 25, 2023 at 12:32 PM Dilip Kumar wrote:
> >
> > > > Is there anything else that stops this patch from supporting migration
> > > > of logical replication slots from PG versions < 17?
that after
commit_flush_lsn no decodable record was generated then we are safe to
upgrade that slot. So this seems an expandable approach.
--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com
701 - 800 of 1730 matches
Mail list logo
