Bug#463763: /etc/init.d/opencryptoki stop does not stop pkcsslotd

2008-07-24 Thread Chaskiel M Grundman
The pkcsslotd.pid file gets the wrong pid, because pkcsslotd forks. So the stop script still doesn't work (and its failure complicates upgrades) whitestar:/var/log# pidof pkcsslotd whitestar:/var/log# cat /var/run/pkcsslotd.pid cat: /var/run/pkcsslotd.pid: No such file or directory whitestar:/va

Bug#463760: opencryptoki: FTBFS if automake1.9 is installed

2008-07-24 Thread Chaskiel M Grundman
It's now failing when automake (=1:1.10.1-3) is installed. I think a better fix for this would be: a) touch aclocal.m4 (and then configure) after the patches are applied, so that make doesn't try to rebuild configure b) make the testcass patch remove the following lines fom Makefile.am, so tha

Bug#476013: owl: FTBFS: configure: error: installation or, configuration problem: C compiler cannot create executables.

2008-05-22 Thread Chaskiel M Grundman
The problem results from this questionable looking line in configure.in FOO=`perl -MExtUtils::Embed -e ldopts | sed 's/,-E//' | sed 's/-liconv//'` It changes '-Wl,-E' to '-Wl' if owl really wants to be built without --export-dynamic, it should remove the -Wl too. -- To UNSUBSCRIBE, email t

Bug#273051: Memory leak in libc getpwuid() and getgrouplist()

2008-05-21 Thread Chaskiel M Grundman
Package: libc6 Version: 2.7-10 Followup-For: Bug #273051 I am seeing this behavior on amd64, both with 2.3.6.ds1-13etch5 and 2.7-10 I think this is a __libc_freeres problem, since if I run valgrind with --run-libc-freeres=no, the relevant blocks are reported as being still reachable (in fact,

Bug#379846: "libsasl2-modules-gssapi-heimdal" isn't anymore (was Re: Bug#379846: Patch for NMU )

2006-09-24 Thread Chaskiel M Grundman
This patch causes libsasl2-modules-gssapi-heimdal to use mit gssapi libraries, becoming equivalent to libsasl2-gssapi-mit. if this build-depends change is the only way to make building cyrus-sasl2 work, then it should probably be arranged that libsasl2-gssapi-mit is built from the cyrus-sasl2

Bug#344543: libkrb53: double free + cache corruption if krb5_get_credentials fails

2005-12-23 Thread Chaskiel M Grundman
Do you have the steps required to duplicate this from a current unstable install? it happens in a newly debootstrap'd sid as soon as I install krb5-user and krb5-clients. answers to the only debconf questions asked are: Name: krb5-config/default_realm Template: krb5-config/default_realm Value:

Bug#305298: ssh-krb5: password authentication does not use pam

2005-04-19 Thread Chaskiel M Grundman
--On Tuesday, April 19, 2005 16:23:28 -0700 Russ Allbery <[EMAIL PROTECTED]> wrote: I'm using ssh-krb5 with PAM here and I don't use that auth mechanism, so I know it does work. Perhaps you're missing UsePAM yes in your sshd configuration? When you log in with a password, do you see [EMAIL PROT

Bug#305298: ssh-krb5: password authentication does not use pam

2005-04-19 Thread Chaskiel M Grundman
--On Tuesday, April 19, 2005 16:23:28 -0700 Russ Allbery <[EMAIL PROTECTED]> wrote: UsePAM yes PasswordAuthentication no -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#295731: ssh fails to bind link-scope IPv6 addresses

2005-04-19 Thread Chaskiel M Grundman
# ssh -6 fe80::2xx:xxff:fexx: ssh: connect to host fe80::2xx:xxff:fexx: port 22: Invalid argument This is not a bug in sshd. It is arguably a bug in the ssh client. Since any given link-local address can potentially exist on all non-loopback interfaces on a host, you need to identify whic

Bug#305298: ssh-krb5: password authentication does not use pam

2005-04-19 Thread Chaskiel M Grundman
--On Tuesday, April 19, 2005 18:22:12 -0400 Sam Hartman <[EMAIL PROTECTED]> wrote: I'd certainly expect pam to be used for all password validation. If that's not true please give me info on how to reproduce. reproduce? As far as I can tell, it never uses pam unless you use the kbdint userauth m