Re: Debugging xfree() (WAS: , in the URL)

Dear Henrik,

as far as I understand, I think that the kind of errors I get from running
"valgrind ./squid -DNYCd3" are not fatal.
I also have the feeling that our changed version of squid is running better
with the -N parameter. The problem is when squid runs without the -N
parameter.

Is there a way to run valgrind against a squid child?
If I don't use the -N parameter, I get no output from valgrind.

cheers,
Mihalis.

----- Original Message -----
From: "Henrik Nordstrom" <hno@squid-cache.org>
To: "Mihalis Tsoukalos" <ekefe@freemail.gr>
Cc: <squid-dev@squid-cache.org>
Sent: Monday, March 24, 2003 11:19 AM
Subject: Re: Debugging xfree() (WAS: , in the URL)

> No, but that there is a buffer overrun, trying to write more data than
> allocated for.
>
> The first chain starting with "at strcpy" tells where the buffer overrun
> occured. The second chain starting with "at malloc" tells where the
> memory was allocated.
>
> What this tells is that on client_side.c:1921 a call to strcpy tried to
> copy more data than allocated for at line 1920.
>
> Regards
> Henrik
>
> Mihalis Tsoukalos wrote:
> >
> > Does the following mean that the malloc() call failed?
> >
> > ==23985==
> > ==23985== Invalid write of size 1
> > ==23985== at 0x4004894C: strcpy (vg_clientfuncs.c:441)
> > ==23985== by 0x8060D28: clientSendMoreData (client_side.c:1921)
> > ==23985== by 0x80A6791: storeClientCallback (store_client.c:165)
> > ==23985== by 0x80A6CFD: storeClientCopy3 (store_client.c:321)
> > ==23985== Address 0x436B9D75 is 0 bytes after a block of size 21
alloc'd
> > ==23985== at 0x40047F4B: malloc (vg_clientfuncs.c:100)
> > ==23985== by 0x80B8570: xmalloc (util.c:439)
> > ==23985== by 0x8060CD9: clientSendMoreData (client_side.c:1920)
> > ==23985== by 0x80A6791: storeClientCallback (store_client.c:165)
> > ==23985==
> >
> > have a nice week,
> > Mihalis.
Received on Mon Mar 24 2003 - 04:49:51 MST