Further to the major memory leaks found by Steinar H. Gunderson
This is also popping out, and very confusing since ICAPXaction global is
only created once right?
==23075==
==23075== 16,245,636 bytes in 155,949 blocks are still reachable in loss
record 29 of 30
==23075==at
On Wed, Feb 11, 2009 at 2:56 AM, Amos Jeffries squ...@treenet.co.nz wrote:
I'm opening this old discussion up again.
[...]
HISTORY:
If I recall correctly, the only holdback we had last time this was
discussed was that certain setups and winbind needed it to work.
That has since changed
On Wed, Feb 11, 2009 at 01:34:23AM +0100, Henrik Nordstrom wrote:
To get an even better trace the best is to build Squid with valgrind
support (--with-valgrind-debug), let it run for a while under valgrind
and then view the memory statistics page.
squidclient mgr:mem
Thanks, that's a neat
Amos Jeffries has voted approve.
Status is now: Semi-approved
For details, see:
http://bundlebuggy.aaronbentley.com/project/squid/request/%3C498FEFF9.1030209%40treenet.co.nz%3E
Project: Squid
allow out-of-tree builds and separate .opts files.
logfile creation moved to test-builds.sh, where it's used.
--verbose mode for showing the actual output while tests is running
test-build.sh generally cleaned up a bit with less duplicated code
# Bazaar merge directive format 2 (Bazaar 0.90)
Bundle Buggy has detected this merge request.
For details, see:
http://bundlebuggy.aaronbentley.com/project/squid/request/%3C1234355448.23043.2.camel%40localhost.localdomain%3E
Project: Squid
bb:approve
--
/kinkie
Hi Amos,
Is there any definitely lost record?
Amos Jeffries wrote:
We seem to have tracked the major leak ( ~1MB per request) down to these:
mem_obj-delayRead(DeferredRead(DeferReader, this, CommRead(fd, buf,
len, callback)));
Which generate:
==21688== 1,251,224 bytes in 12,031
Hi Amos,
Amos Jeffries wrote:
Further to the major memory leaks found by Steinar H. Gunderson
This is also popping out, and very confusing since ICAPXaction global is
only created once right?
it is the new PconnPool line. It is create only once. It says its
reachable because the time
May I ask, about the votes, which is the procedure? Who is giving access
to bundlebuggy? Or it is something I do not need it?
On Wed, 2009-02-11 at 23:20 +0200, Tsantilas Christos wrote:
May I ask, about the votes, which is the procedure? Who is giving access
to bundlebuggy? Or it is something I do not need it?
I can add users, but I don't consider myself the decider - rather the
community here is.
-Rob
ons 2009-02-11 klockan 14:56 +1300 skrev Amos Jeffries:
WHY:
* it's a security breach.
Why?
* it's the source of many permissions annoyances.
Yes.
* the setting is still widely recommended in online how-to's
Yes, and often for the wrong reasons.
* current Squid-3+ are perfectly
Robert Collins wrote:
On Wed, 2009-02-11 at 23:20 +0200, Tsantilas Christos wrote:
May I ask, about the votes, which is the procedure? Who is giving access
to bundlebuggy? Or it is something I do not need it?
I can add users, but I don't consider myself the decider - rather the
community here
tor 2009-02-12 klockan 08:25 +1100 skrev Robert Collins:
On Wed, 2009-02-11 at 23:20 +0200, Tsantilas Christos wrote:
May I ask, about the votes, which is the procedure? Who is giving access
to bundlebuggy? Or it is something I do not need it?
I can add users, but I don't consider myself
tor 2009-02-12 klockan 08:25 +1100 skrev Robert Collins:
On Wed, 2009-02-11 at 23:20 +0200, Tsantilas Christos wrote:
May I ask, about the votes, which is the procedure? Who is giving
access
to bundlebuggy? Or it is something I do not need it?
I can add users, but I don't consider myself
ons 2009-02-11 klockan 14:56 +1300 skrev Amos Jeffries:
WHY:
* it's a security breach.
Why?
Overriding the underlying OS, which admin may understand, with behavior
they may not. Can cause them to enact less secure workarounds; I have
seen squid effective-user'd to the root UID not long
tor 2009-02-12 klockan 12:30 +1300 skrev Amos Jeffries:
Overriding the underlying OS, which admin may understand, with behavior
they may not. Can cause them to enact less secure workarounds; I have
seen squid effective-user'd to the root UID not long ago.
cache_effective_user root is not
Hi Amos,
Amos Jeffries wrote:
Further to the major memory leaks found by Steinar H. Gunderson
This is also popping out, and very confusing since ICAPXaction global is
only created once right?
it is the new PconnPool line. It is create only once. It says its
reachable because the
Hi Amos,
Is there any definitely lost record?
Only minor stuff which isn't memPool'd
Amos Jeffries wrote:
We seem to have tracked the major leak ( ~1MB per request) down to
these:
mem_obj-delayRead(DeferredRead(DeferReader, this, CommRead(fd, buf,
len, callback)));
Which
On 02/11/2009 06:08 PM, Amos Jeffries wrote:
Is there any definitely lost record?
Only minor stuff which isn't memPool'd
Is it possible that indirectly lost is not really lost?
Thank you,
Alex.
Amos Jeffries wrote:
We seem to have tracked the major leak ( ~1MB per
On 02/11/2009 06:08 PM, Amos Jeffries wrote:
Is there any definitely lost record?
Only minor stuff which isn't memPool'd
Is it possible that indirectly lost is not really lost?
No. From the valgrind manual on direct vs indirect leaks:
The distinction is that a direct leak is a block
21 matches
Mail list logo
