Tom Lane wrote:
> I'd accept a mechanism to enforce a timeout at the lock level if you
> could show me a convincing use-case for lock timeouts instead of
> statement timeouts, but I don't believe there is one. I think this
> proposal is a solution in search of a problem.
I think statement_timeout
Chris KL just raised an issue on IRC:
test=> create table test (a int4) tablespace pg_default;
ERROR: permission denied for tablespace pg_default
This wasn't encountered in my original patch because
pg_tablespace_aclmask() had this test reasonably early on:
+ if(tbloid == DEFAULTTBLSPC)
+
Satoshi Nagayasu <[EMAIL PROTECTED]> writes:
> statement_timeout terminates large sort or scan
> even if it is running, doesn't it?
> statement_timeout doesn't care that
> the process is waiting a lock or running.
> I don't want to terminate a running query.
> So a lock waiting backend shold be k
statement_timeout terminates large sort or scan
even if it is running, doesn't it?
statement_timeout doesn't care that
the process is waiting a lock or running.
I don't want to terminate a running query.
So a lock waiting backend shold be killed.
Tom Lane wrote:
> Satoshi Nagayasu <[EMAIL PROTEC
Satoshi Nagayasu <[EMAIL PROTECTED]> writes:
> I guess the transaction cancellation from the client
> using PQrequestCancel() is available, but the cancellation
> logic must be implemented in the client-application using
> signal or thread.
Actually I think the recommended solution involves using
Tom,
I guess the transaction cancellation from the client
using PQrequestCancel() is available, but the cancellation
logic must be implemented in the client-application using
signal or thread.
I think detecting such situation on server-side is not
available now, and SQL Server or DB2 have same fu
Satoshi Nagayasu <[EMAIL PROTECTED]> writes:
> When a transaction is blocked by another transaction because of
> waiting a lock, we need a lock timeout in some cases.
Isn't there an existing solution for this problem?
regards, tom lane
---(end of b
Yeah, that's still on my should-fix-for-7.5 list (and I think Fabien was
going to, or already did, submit some ACL-hacking code to help). That
is, ALTER OWNER should adjust the ACL so that grants made by/to the
former owner now appear to be by/to the new owner.
However, there's still the problem t
Hello all,
I've created a lock timeout patch and it's attached.
When a transaction is blocked by another transaction because of
waiting a lock, we need a lock timeout in some cases.
Using this patch, the lock timeout is enabled with
'lock_timeout = ' directive in postgresql.conf,
and if a ti
Christopher Kings-Lynne <[EMAIL PROTECTED]> writes:
> ... The acl is still there from when brett used to own that table?
> Do you still plan to fix that?
Yeah, that's still on my should-fix-for-7.5 list (and I think Fabien was
going to, or already did, submit some ACL-hacking code to help). That
I've been bothered for awhile about a couple of inconsistencies in our
handling of user-defined type names: you can't schema-qualify a type
name that you use to prefix a literal constant, and you can't use
typmod qualifiers with user-defined types. Shachar Shemesh's complaint
today about the latte
* Drop commands for TYPEs have 'CASCADE' on the end (has that always
been true)
Yeek. That's got to be a hangover from pre-dependency-chasing days.
Let's lose it in our current output, at least.
I think it's necessary due to the circular dependency between types and
their I/O functions.
Chris
-
I don't buy it. There's a tradeoff here between certainty of doing what
you want and having a script that is easy to edit. DROP is a dangerous
weapon and we should be circumspect about applying it, but ALTER OWNER
etc are much less so.
Also, the point about qualifying the DROP is that you do not
Christopher Kings-Lynne <[EMAIL PROTECTED]> writes:
> We currently fully qualify DROP command with the namespace so that drops
> will not accidentally modify the system catalogs. Shouldn't this also
> be necessary on ALL non-CREATE commands?
> Otherwise, if the create table command associated wi
Christopher Kings-Lynne <[EMAIL PROTECTED]> writes:
> Actually, this brings up another point - people occasionally complain on
> the list that pg_dump is not considered important enough :( ie. Is
> there any good reason we cannot backport the entire new pg_dump to the
> 7.4 branch, and change t
* Do we no longer worry about the SCHEMA AUTHORIZATION clause? I might
set it to keep being issued in 'sql standard mode', but otherwise we
cannot use it in dumps any more.
Actually, that's not true - I'm being silly. We can use the
AUTHORIZATION clause instead of ALTER SCHEMA ... OWNER TO :)
Ugh. Definitely an oversight. Don't suppose you want to think about
pulling the name out of the DROP command ;-) ?
Yeah, I've already done it - it's ugleeey, but it works :P
I'm running out of time unfortunately, and I need to know from you
whether I should go back to my work on making owner and
Shachar Shemesh wrote:
> What do I need to do in order to get the width specifier into my
> type?
This is not possible with user-defined types.
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
Shachar Shemesh <[EMAIL PROTECTED]> writes:
> What do I need to do in order to get the width specifier into my type?
Rewrite the grammar. Width modifiers are only supported on types that
are hard-wired into the grammar, mainly because they look way too much
like function calls to be distinguishe
Christopher Kings-Lynne <[EMAIL PROTECTED]> writes:
> I tried adding the extra scan in and it as all well and good up until
> the second where I realised that the TocEntry struct has no field that
> allows me to know the correct way of finding the full descriptor of each
> object.
Ugh. Definit
Hi list,
I'm trying to create a varchar clone (called varcharci). I have defined
new functions called "varcharciin", "varcharciout", "varcharcisend" and
"recv", using the "varcharin" etc. definitions (i.e. - builtin
functions). I defined the type. Everything works, except that when I try
to cre
But...it seems kind of hacky to scan it again for owners and privs - are
you sure you want me to go that way?
If there's not a big performance penalty, sure. Being fully compatible
with existing archive files is a sufficient win to justify sins much
worse than this one.
Ah, crap.
I tried adding t
22 matches
Mail list logo
