On Mon, May 28, 2018, 02:00 Erwin Brandstetter wrote:
> I found an existing bug report and have something to add to it.
>
> What's the best way to reply to it? Just using a browser, with no
> newsreader installed.
>
> This one:
>
> https://www.postgresql.org/message-id/flat/20170925084522.1442.32
ave decent coverage of other cultures and countries as well. We can't
cover them all on the committee (that would make for a gicantic
committee), but we can cover it with people who are used to communicating
and working with people from other areas as well, which makes for a better
understanding.
It won't be perfect in the first attempt, of course, but that one is
covered.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Tue, Jun 26, 2018 at 11:40 AM, Ivar Fredriksen wrote:
> A single large query is able to spend all the system memory (as reported
> by top), and the oom_reaper kills postgres. See bottom of email for an
> example query and logs.
>
>
>
> Expected behavior would be that postgres is not killed and
something like
this again, you will be banned immediately and permanently from the lists.
--
Magnus Hagander
PostgreSQL Core Team
t to the new
location, and start it up again.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
Jul 6, 2018 at 5:49 PM Magnus Hagander
> wrote:
>
>> On Fri, Jul 6, 2018 at 6:42 PM, Duarte Carreira
>> wrote:
>>
>>> Hello.
>>>
>>> Yes I'm one of those guys who only recently realized the mess of having
>>> tablespaces inside t
ol from others, but it's not exactly news... But they do
outline a very definite problem, which is that if you get physical
corruption in your database, it gets included in the backups. And if it's
in a portion of the database you don't use a lot, checksum failures won'
that the project could
*start out* using the wiki, and once there is enough content to prove the
idea other platforms could be looked at and it would be easy enough to
migrate that data out there (even if just by copy/paste) if it becomes a
need.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
up version: (PostgreSQL) 10.10
>
>
>
> Anybody else encountering this problem?
>
>
>
I think your problem is that you are looking at the exit code from "tee"
and not from pg_basebackup. If you are using bash, you can look at
something like $PIPESTATUS to get the exit
u want to get an individual security patch you will have to cherry
pick it from git and build your own server from source. But per the above
link, it is really recommended that you don't do that. Instead, do it the
way it's intended to, which means install the latest minor release.
Why
n still has to actually try to
log in with lowercase, and do so before it connects to PostgreSQL.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Fri, Aug 30, 2019 at 3:00 PM Niels Jespersen wrote:
> Hello Magnus
>
> Thank you for your prompt reply.
>
> I’m not sure I understand your last statement. I want to achieve that
> regardless of the case of the entered username is logged into the same
> Postgres user (whose name is created in a
on. So we obviously have to wait for EDB to provide updated lists there --
but until then, you can probably expect it to work just fine.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
, but if not you should be complaining to that vendor that they
are missing important fixes, rather than try to install a version with
known bugs and security issues.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Thu, Nov 21, 2019 at 1:46 PM stan wrote:
> I am trying to set up to do some work with pg_dump, and I would like to be
> able to connect from my normal user to do this. This is on a Ubunt 18.04
> installation. I have added the follwing to pg_hba.conf:
>
> hostall all
On Fri, Dec 6, 2019 at 10:50 AM Zwettler Markus (OIZ) <
markus.zwett...@zuerich.ch> wrote:
> > -Ursprüngliche Nachricht-
> > Von: Michael Paquier
> > Gesendet: Freitag, 6. Dezember 2019 02:43
> > An: Zwettler Markus (OIZ)
> > Cc: Stephen Frost ;
> pgsql-general@lists.postgresql.org
> > B
ase the archiving only has to wait for it to acknowledge
the process has started, not finished.
There's always a risk involved in returning from archive_command before the
file is safely stored on a different machine/storage somewhere. The more
async you make it the bigger that risk is, but it increases your ability to
parallelize.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
e not in-order according
> to the new behavior of the text comparison operators, leading to havoc
> because btree searching relies on the entries being correctly sorted.
See https://wiki.postgresql.org/wiki/Locale_data_changes for hints on
which linux distros updated when.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
isk immediately. If you are doing compression it uses a little more,
but still within the "you don't really need to care" range. The RAM
usage is not dependent on the size of the database, it is always that
small.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
accumulated sizable data in your cluster. (Turning it on in this
case is easy, but not fast).
And FWIW, I do think we should change the default. And maybe spend some
extra effort on the message coming out of pg_upgrade in this case to make
it clear to people what their options are and exactly wh
directly return utc.
SELECT read_time AT TIME ZONE 'utc'
will return the time in UTC (as a timestamp)
And just make sure you have done a "set time zone 'utc'" before you *load*
the data, and everything should just work automatically.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Thu, Apr 16, 2020 at 6:12 AM Niels Jespersen wrote:
>
>
>
>
> *Fra:* Magnus Hagander
> *Sendt:* 15. april 2020 20:05
> *Til:* Niels Jespersen
> *Cc:* pgsql-general@lists.postgresql.org
> *Emne:* Re: timestamp and timestamptz
>
>
>
>
>
>
>
&
On Sun, Apr 12, 2020 at 4:23 PM Tom Lane wrote:
> Magnus Hagander writes:
> > And FWIW, I do think we should change the default. And maybe spend some
> > extra effort on the message coming out of pg_upgrade in this case to make
> > it clear to people what their options ar
ackup of some sort just before shutting down the server.
>
> Yes indeed!
>
Also don't forget that on an upgrade from Debian 9 to 10, you most likely
need to reindex your string indexes, see
https://wiki.postgresql.org/wiki/Locale_data_changes
<https://wiki.postgresql.org/wiki
Hello!
This address is for reporting issues with the PostgreSQL website, it is not
a help desk. For general support questions, please see
https://www.postgresql.org/support/ - in particular, consider the
pgsql-general public mailinglist, as you seem to have already done, and
received a reply there
tgreSQL these days,
one should be using scram-sha-256 which does not have this problem (and has
been around for a few years by now)., if using local database logins.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
PQnotifies()
once that call has indicated "something happened".
Similar applies to writing such daemons using for example the python or
perl interfaces. You block your process or thread on select() and take
action when that one returns.
--
Magnus Hagander
Me: https://www.hagand
On Thu, May 14, 2020 at 9:01 AM Santhosh Kumar
wrote:
> Hi,
> I came across a link published in postgresql, where it is clearly
> mentioned BDR as an open source. When I tried to install BDR for CentOS
> from 2ndQuadrant, the yum repository was not reachable and upon further
> enquiring with 2nd
lot
2. pg_basebackup with slot
3. start replication with slot
4. restart replication without slot once it's caught up
5. drop slot
However, if you want reliable replication, you really should have a slot.
Or at least, you should have either a slot *or* log archiving that's
read-accessible from the replica.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
gt;
All the PostgreSQL versions available om yum from the postgresql.org site
have SSL enabled. Just install using the instructions from
https://www.postgresql.org/download/.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
s change in its development, but it does
> ! provide page-level,
>
> Ah, well, anyway that seems to be something significantly smaller
> than the usual 1 gig table file at once.
>
pg_probackup does page level incremental *if* you install a postgres
extension that some people hav
ntly I know so much:
> - it writes a backup_label file. That is just a few lines of
>ASCII and should not be difficult to produce.
>
It does that only in exclusive mode, and doing that is one of the big
problems with exclusive mode. So don't do that.
> I now hope ver
On Sat, Jun 13, 2020 at 10:13 PM Peter wrote:
> On Thu, Jun 11, 2020 at 10:35:13PM +0200, Magnus Hagander wrote:
> ! > Okay. So lets behave like professional people and figure how that
> ! > can be achieved:
> ! > At first, we drop that WAL requirement, because with WAL
eral definitely needs improvement,
this particular requirement is documented at
https://www.postgresql.org/docs/current/backup-file.html.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
> concurrent backup capabilities should yield any benefit at all.
>
One not uncommon case is for example being able to provision a new replica
while a backup is running. Since replicas are provisioned starting off a
base backup, being able to run that concurrently is very useful. Especia
e the
> database was running in backup mode during which the crash occurred. How
> does that work if no pg_stop_backup() output exists? Did I miss something
> here?
>
It does not work off *that* base backup. But if you start from the *prior*
be backup (one that did complete wit
e actual autovacuum
details -- can you include those? That is, you are only including the very
last row of the log message, but the interesting parts are at the beginning.
I assume you've also looked for other autovacuum messages in the log --
such as it being canceled by concurrent activity?
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
ERS).
You are trying to create columns with names that are way longer than that,
so they will be truncated. If you create this table in psql it will show
you a NOTICE information about this -- I believe in psycopg2 this shows up
in conn.noticies (where conn is the connection object
sistency group will set up a write
fence across them, then snapshot, and AIUI guarantees correct write
ordering.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
; Or even if you come into a place to support an existing PostgreSQL
> database, how do you find out what the database segsize is?
>
>
You can run the query "SHOW segment_size" to show the compiled-in value.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.
mmands can alter the data. To make sure that's not what's happening, you
may want to try doing the same thing with a BEGIN TRANSACTION ISOLATION
LEVEL SERIALIZABLE instead, and see if the problem still occurs.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
is probably why the package version has a
name ending in "sat". So it would be something a Linux admin would put in
there, not the DBA.
But to answer the question, no they are not required by PostgreSQL, repmgr
or pgbouncer.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
, varchar just takes a number, not the special construct with
BYTE. PostgreSQL varchar:s always limit the size based on number of
characters, not bytes.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Thu, Sep 3, 2020 at 12:43 PM Hou, Zhijie
wrote:
> Hi all
>
> In master branch, I found a typo in Comments of function
> HandleProcSignalBarrierInterrupt.
> See the attachment for the patch.
>
>
Thanks, pushed.
//Magnus
(Please don't drop the mailinglist from CC, as others are likely interested
in the responses)
On Tue, Sep 8, 2020 at 3:06 PM Michael Holzman
wrote:
>
>
> On Tue, Sep 8, 2020 at 3:03 PM Magnus Hagander wrote:
>
>> A PostgreSQL SELECT does *not* open a transaction past the
On Tue, Sep 8, 2020 at 4:01 PM Michael Holzman
wrote:
>
>
> On Tue, Sep 8, 2020 at 4:25 PM Magnus Hagander wrote:
>
>>
>>
>> Whether you have autocommit on or off, you can *always* control things
>> explicitly. And you can certainly run "multi-statemen
On Tue, Sep 8, 2020 at 4:38 PM Tom Lane wrote:
> Magnus Hagander writes:
> > Oh sure, but there is clearly *something* going on, so we should try to
> > figure that out. Because a transaction running multiple independent
> selects
> > with the defaults settings
On Tue, Sep 8, 2020 at 5:16 PM Tom Lane wrote:
> Magnus Hagander writes:
> > On Tue, Sep 8, 2020 at 4:38 PM Tom Lane wrote:
> >> The reason that's not so is that whether or not transaction A *has*
> >> touched table B is irrelevant. It *could* read tab
and all should be taken care of.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
ve caused corruption elsewhere as well, so
whatever verification you can do against other tables, you should do as
well.
You'll of course also want to check any kernel logs or storage system logs
to see if they can give you a hint as to what happened, but they are
unlikely to ac
s the same.
Once it fails, you've found a corrupt block...
//Magnus
On Tue, Sep 15, 2020 at 12:46 PM Vasu Madhineni
wrote:
> Is it possible to identify which rows are corrupted in particular tables.
>
> On Tue, Sep 15, 2020 at 5:36 PM Magnus Hagander
> wrote:
>
>>
&g
urn
rows in a consistent/predictable order. So as long as that query is part of
what you're doing, you should not be surprised if you get the rows in an
inconsistent/unpredictable order, with whatever follow-on effects that
might have. (And it can lead to weird follow-on effects like the ones
y
adhineni
wrote:
> I could see block read I/O errors in /var/log/syslog. if those error fixed
> by OS team, will it require recovery.
>
> Also can i use LIMIT and OFFSET to locate corrupted rows?
>
> Thanks in advance
>
> Regards,
> Vasu Madhineni
>
> On Wed, Sep
> Was it?
>
> Pre-10 it was:
>
> MAJOR.MAJOR.PATCH
>
Yeah the fact that it kind of looked like semver, but *wasn't* semver, is
probably one of the (many) things that confused people. It definitely
wasn't semver.
--
Magnus Hagander
Me: https://www.hagande
shared_buffers';
>
> Ah, forgot that shared_buffers is in 8K pages.
>
> So you actually need:
>
>select pg_size_bytes(setting) * 8192
>from pg_settings
>where name = 'shared_buffers';
>
Actually, it doesn't have to be in 8k pages, that d
On Wed, Oct 14, 2020 at 5:10 PM Tom Lane wrote:
> Magnus Hagander writes:
> > On Wed, Oct 14, 2020 at 3:57 PM Thomas Kellerer wrote:
> >> select pg_size_bytes(setting) * 8192
> >> from pg_settings
> >> where name = 'shared_buffers';
>
>
On Wed, Oct 14, 2020 at 5:23 PM Tom Lane wrote:
> Magnus Hagander writes:
> > On Wed, Oct 14, 2020 at 5:10 PM Tom Lane wrote:
> >> It's fairly annoying that this doesn't work:
> >> regression=# select pg_size_bytes(setting||' '||unit) fro
Yes, it is free.
> PostgreSQL(9.4 - 11) support subscription
>
The PostgreSQL community does not provide subscriptions. You can find
information about the support options, both free and paid, at
https://www.postgresql.org/support/.
--
Magnus Hagander
Me: https://www.hagande
hasn't made
> production yet.
>
PostgreSQL does not have such a thing as "open feature requests".
You can find patches that are currently being worked on at
https://commitfest.postgresql.org/, or on discussions in the list archives.
--
Magnus Hagander
Me: https://www.
course have to figure out how to securely authenticate the postgres
OS user on the standby node to the Kerberos system, but that's doable.
(Though I believe most Kerberos implementations also rely on filesystem
security to protect the tickets, so if you don't trust your filesystem, you
may have a problem with that -- as well as indeed most other authentication
systems -- so you'd have to investigate that within the kerberos system).
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
onsibility over to your Kerberos system. Then you can set that one
up to require you to manually type in a password or equivalent to get t a
ticket, and configure expiry on that ticket.
//Magnus
*发件人:* Magnus Hagander [mailto:mag...@hagander.net]
>
> *发送时间:* 2020年10月27日 17:00
> *收件
at
https://www.postgresql.org/docs/13/server-start.html talks about this
behaviour, but only notes that crash recovery might be a reason to hit
this timeout. Maybe it needs to also mention replication (and probably
archive recovery)?
> The best place to discuss this would be the "pgsql-pkg-yum" list.
I don't think this is a packaging issue, all the RPMs did was enable
the functionality that's in core postgresql.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
m before handing the actual call up to the operating system. It's
completely independent of how the file is opened.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
ore muscles": "No",
> "Have you travelled outside of Canada in the past 14 days?": "No",
> "Have you had close contact with a confirmed or probable case of
> COVID-19?": "No"
> }
>
> If the order had remained the same, it's child's play to pull the data out
> and present it in a report, even if the data elements change.
> But... seen above, the order gets mixed up.
>
> Any ideas?
The json standard declares that the keys in a document are unordered,
and can appear at any order.
In PostgreSQL, jsonb will not preserve key ordering, as a feature for
efficiency. The plain json datatype will, so if key ordering is
important you should use json instead of jsonb (but you should
probably also not use the json format in general, as it does not
guarantee this)
See https://www.postgresql.org/docs/13/datatype-json.html
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
it, like
SELECT "cast" FROM test
(and the same when you create the table, or indeed any references to the column)
zone is not, and should be fine.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
ple this thread here
https://www.postgresql.org/message-id/flat/61463e206b7c4c0ca17b03a59e890b78%40lmco.com,
and the config on https://github.com/rc9000/postgres-fail2ban-lockout.
(probably needs some small adaptations, but as a base it should work).
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
:
dnf install --excludepkg proj --excludepkg proj-datumgrid postgis30_12
as a workaround.
*If* the root cause is the same one, that is...
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
On Fri, Nov 13, 2020 at 7:10 PM Magnus Hagander wrote:
> dnf install --excludepkg proj --excludepkg proj-datumgrid postgis30_12
> postgis30_12-devel postgis30_12-utils postgis30_12-client postgis30_12-docs
>
> On Fri, Nov 13, 2020 at 7:01 PM Tom Lane wrote:
> >
> &g
endent objects too.
If you have no functions using it, it will just go away, and once you
have dropped it in both databases you should be good to go.
And of course, if there are functions depending on it, you should
rebuild those on plpython3u before you drop plpython2u (or drop the
functions if they're not in use).
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
mostly
fine on RHEL7. But if you don't actually show us what your dependency
problems are, we can't tell you how to fix it...
(And why not use Patroni from the PDGD repositories?)
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
mand really not have a %f in it anywhere? That
definitely seems wrong... But it does seem to copy some files correct,
which would be weird if it doesn't. Mistake in the report, or is there
something really weird going on with that PostgreSQL_DEV not being a
directory but instead some "magic file"?
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
specifically looking at the Debian or Ubuntu packages, you can
find the full packaging information in the salsa repositories at
https://salsa.debian.org/postgresql/postgresql. It will have all teh
details you need.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/&g
1.3
when you use the old version...
I assume you're running both the 11 and the 13 client on the same host?
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
n.nnn.n.nnn", user
> "kalle", database "postgres", SSL off
> FATAL: no pg_hba.conf entry for host "nn.nnn.n.nnn", user "kalle", database
> "postgres", SSL off
>
> KR Mikael Gustavsson, SMHI
>
>
> ___
sql.org/
If not, then you will have to build from source manually -- the old
versions of PostgreSQL are still available in source form on
https://ftp.postgresql.org/pub/source/
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
bonus you get Kerberos which is a lot more
secure than ldap for auth.. It might have a slightly higher barrier
of entry, but could probably pay off well in a case like this.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
On Wed, Jan 6, 2021 at 4:39 PM Paul Förster wrote:
>
> Hi Magnus,
>
> > On 06. Jan, 2021, at 15:48, Magnus Hagander wrote:
> >
> > Only if you can create rules in your pg_hba.conf file that knows where
> > the users are. You can specify multiple servers on one
ntence -- without that one, your
reading of it would make more sense. See also the following parameter,
ldaptls, which uses similar language.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
On Thu, Jan 7, 2021 at 10:40 AM Paul Förster wrote:
>
> Hi Magnus,
>
> > On 06. Jan, 2021, at 16:57, Magnus Hagander wrote:
> >
> > Yes. But you have a really hacky environment :P
>
> actually not. We have an old LDAP which we want to retire this year. And
cluded your problem seems to be with the
proprietary server product from EnterpriseDB, not with PostgreSQL.
For support with that, you should contact the EDB support channels.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
hing, and
then reply to that through your normal email. It's not the most
convenient workflow, but if you mostly read and only very seldom post,
it works.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
ges
Note that this is triggered by the Ubuntu upgrade, not by upgrading
PostgreSQL -- that's why it happend even when you keep using the same
PostgreSQL version.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
ing about it).
Buttom line is that while it may be a good tool for reading, it is
*not* a good tool for posting, at least not until they fix their basic
handling of email.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
dy from the mailing list admins help me out here?
Hi!
You seem to have hit a bug in our oauth1 processing (twitter is the
only provider we use oauth1 for, the others use oauth2). I believe
it's been fixed now, please try again.
--
Magnus Hagander
Me: https://www.hagander.net/
Work: https://www.redpill-linpro.com/
obviously I understand creating many indexes will impact write
> performance and space will be utilized.
>
> Would like to know if creating indexes in this manner can create any other
> issues or inputs on the whole topic of indexes in JSONB types.
>
You should not be creating i
hs away.
You definitely shouldn't be using that.
Buster is also considered the "oldstable" version. You should probably be
using bullseye. (I assume you're talking about some generic binaries and
not the DEB packages of course -- DEB packages should be built on their
corres
On Fri, Apr 1, 2022 at 4:52 PM Daniele Varrazzo
wrote:
> On Fri, 1 Apr 2022 at 16:28, Magnus Hagander wrote:
>
> > Stretch also goes end of life on June 30 2022, so just a few months
> away. You definitely shouldn't be using that.
>
> The platform is part of the Py
On Fri, Apr 1, 2022 at 5:08 PM Daniele Varrazzo
wrote:
> On Fri, 1 Apr 2022 at 17:00, Magnus Hagander wrote:
> >
> > On Fri, Apr 1, 2022 at 4:52 PM Daniele Varrazzo <
> daniele.varra...@gmail.com> wrote:
> >>
> >> On Fri, 1 Apr 2022 at 16:28, Magnus H
new database from initdb (I would also say create a
completely new filesystem underneath it since that's where the corruption
is, if that's easily done)
4. Restore the pg_dump. At this point it will throw errors on any foreign
keys that are "off", and you will have to c
user permissions, none of which it's
recommended that the application run with. Doesn't help once the problem
has occurred of course, but can help avoid it happening in the future.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
a standby server, and there is replication lag
>
There's also:
c) The SELECT runs in a transaction stat *started* before the transaction
that a runs in. (Assuming it then retries with a new transaction later,
that is)
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Fri, Apr 8, 2022 at 3:07 PM Jan Wieck wrote:
> On 4/8/22 08:58, Magnus Hagander wrote:
> > A side-note on this, which of course won't help the OP at this point,
> > but if the general best practice of not running the application with a
> > highly privileged accoun
ill
fail to commit if the foreign key is broken *at that point*. But it lets
you do things like modify multiple tables that refer to each other, and
have the changes only checked when they're all done.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Fri, Apr 8, 2022 at 3:27 PM Magnus Hagander wrote:
>
>
> On Fri, Apr 8, 2022 at 3:23 PM Perry Smith wrote:
>
>>
>>
>> On Apr 8, 2022, at 07:47, Jan Wieck wrote:
>>
>> On 4/8/22 01:57, Nikolay Samokhvalov wrote:
>>
>> On Thu, Apr 7, 2022
and it worked
fine from all of them, even when artificially slowing it down to something
much slower.
So yes, there is *something*, but it's not with in the pg.org
infrastructure.
One thing we got to work that time, I think, was to run:
git config --global http.version HTTP/1.1
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
ified that the new one works. This looks
like a debian/ubuntu system, which means you probably forgot to run
"pg_dropcluster 12 main"? Or if it's not a debian cluster, the equivalent
of that which results in removing the data directory for 12 along with any
configuration files
ity?
>
>
Are these two really running on the same operating system?
This looks a lot like the locale changes included in newer versions of
glibc, and is in that case dependent on an upgrade of the operating system,
not an upgrade of PostgreSQL. See
https://wiki.postgresql.org/wiki/Locale_data_changes for details.
--
Magnus Hagander
Me: https://www.hagander.net/ <http://www.hagander.net/>
Work: https://www.redpill-linpro.com/ <http://www.redpill-linpro.com/>
On Mon, Jun 27, 2022 at 12:01 PM Laurenz Albe
wrote:
> On Mon, 2022-06-27 at 11:38 +0200, Magnus Hagander wrote:
> > On Mon, Jun 27, 2022 at 11:30 AM Florents Tselai <
> florents.tse...@gmail.com> wrote:
> > > A few months back (October) I had upgraded a Postgre
f risk
> taken when snapshotting a PGSQL instance spread across two (or more)
> different pools?
>
>
>
"Don't do it".
If you can't get atomic snapshots, don't do it, period.
You can use them together with a regular online backup. That is
pg_start_backup(
On Sun, Jan 15, 2023 at 10:57 PM HECTOR INGERTO
wrote:
>
>
> > But you cannot and should not rely on snapshots alone
>
>
>
> That’s only for non atomic (multiple pools) snapshots. Isn’t?
>
Right. For single-filesystem installs it should be fine. Just make sure it
has both the data and the WAL di
1 - 100 of 181 matches
Mail list logo
