Bug#307626: samba: Erroneous permissions checks after 3.0.10 -> 3.0.14a
Package: samba Version: 3.0.14a-1 Severity: important The situation: Samba share with 'force group' directive, directories with write access for that group. After upgrade to 3.0.14a strange things start happening: mv Dir1/file Dir2/ - access denied! mkdir Dir1/tmp - OK touch Dir1/tmp/file1 - OK mv Dir1/tmp/file1 Dir1/ - OK mv Dir1/file1 Dir1/tmp/ - access denied! downgrading to 3.0.10 brings situation back to normal. -- System Information: Debian Release: 3.1 APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i586) Kernel: Linux 2.4.30-bsd29a Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Versions of packages samba depends on: ii debconf [debconf-2.0] 1.4.30.13Debian configuration management sy ii libacl1 2.2.23-1 Access control list shared library ii libattr12.4.16-1 Extended attribute shared library ii libc6 2.3.2.ds1-21 GNU C Library: Shared libraries an ii libcomerr2 1.37-2 common error description library ii libcupsys2-gnutls10 1.1.23-10Common UNIX Printing System(tm) - ii libkrb531.3.6-2 MIT Kerberos runtime libraries ii libldap22.1.30-3 OpenLDAP libraries ii libpam-modules 0.76-22 Pluggable Authentication Modules f ii libpam-runtime 0.76-22 Runtime support for the PAM librar ii libpam0g0.76-22 Pluggable Authentication Modules l ii libpopt01.7-5lib for parsing cmdline parameters ii logrotate 3.7-2Log rotation utility ii netbase 4.21 Basic TCP/IP networking system ii samba-common3.0.14a-1Samba common files used by both th -- debconf information excluded -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#147280: Assigning UIDs over 65536 should not be a bug anymore
> tags 147280 wontfix > retitle 147280 [TO CLOSE 20050405] passwd: newusers finds max userid as base > - on debian there is nobody user with 65534 uid > thanks > > As far as I know, assigning UIDs over 65534 is no more a problem, so I > suppose that the bug you reported should now be closed. > Do you have any objection to this? I think that the original problem, with treating max userid as base, should be fixed. One can easily imagine someone creating uid==2**32, besides such behaviour is a bit unelegant. It should search for lowest available userid, or at least this behaviour should be configurable. regards, Eyck -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 We're giving you a new chance in life, and an opportunity to screw it up in a new, original way. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#329773: cpqarrayd: Severely leaks memory ( 1G after few months of working )
> Which version of cpqarrayd are you using? (the bug report doesn't say) latest stable - 2.0-3, > Some memory changes to clean things up were added in the 2.0-4 version, if I am testing 2.0-4 right now, no signs of leakage yet, I do believe that this change should be commited to stable, I was very surprised to find my servers die, while there are known fixes available. -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 Total Existance Failure -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#329773: cpqarrayd: Severely leaks memory ( 1G after few months of working )
Package: cpqarrayd Severity: normal All cpqarrayds running on Proliant 380 G2+ are leaking memory, slower on machines with only one builtin controller (0.5G after few months), and scarily fast on machines with more then one controller. -- System Information: Debian Release: 3.1 APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i686) Kernel: Linux 2.4.28-bsd25d Locale: LANG=C, LC_CTYPE=pl_PL (charmap=ISO-8859-2) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#195408: acknowledged by developer (libio-socket-ssl-perl: I'd like to be able to switch ssl on on existing IO::Socket::INET object)
> the solution Peter posted seems to do what you want. Therefor I close > this bug now. Correct, this method was added in 0.93, and as 0.96 is already in debian, it's all good. Thanks. -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 Total Existance Failure -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#400103: closed by Jose Parrella <[EMAIL PROTECTED]> (Bug#400103: fixed in nginx 0.4.13-2)
Just a little followup as a warning for people caught in this: upgrading with this fixed nginx version removes your carefully re-created /var/www/index.html ;) -- Dariush Pietrzak, Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#400103: closed by Jose Parrella <[EMAIL PROTECTED]> (Bug#400103: fixed in nginx 0.4.13-2)
> > upgrading with this fixed nginx version removes your carefully re-created > Dariush: > > Do you mean that the bug is still present in 0.4.13-2? Here [1] is a As far as I know, 0.4.13-2 is OK, however, when I upgrade from 0.4.13-1 (which overwrote my index.html) to -2, the process removes those files from /var/www: index.html 50x.html, which is kind of bad..., this is just a note to those caught by this buglet, for anyone else, this is irrelevent. -- Dariush Pietrzak, Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#290848: openwebmail: Friviolous use of suid bit. Possible remote root.
Package: openwebmail Version: 2.41-6 Severity: important Tags: security Quote: " OpenWebmail needs suid. Setting... Done. Initializing. It could take a while... ". This happens after update, while documentation states: "...and you want to improve the security of your system ... ... 4. Change permissions chmod -s /usr/share/openwebmail/cgi-bin/*.pl " This leads to package unexpectedly running suid root. And those are complicated pieces of unaudited code, running even without "-T". This setup leads to spellchecker and calendar running as root behind the users/admins back. Documentation should be fixed to state that dpkg-statoverride should be used AND postinst shouldn't ignore existing statoverride ie: for f in `dir -1 /usr/lib/cgi-bin/openwebmail/openwebmail*.pl`; do if [ "`dpkg-statoverride --list $f`" ]; then #don't touch existing override! dpkg-statoverride --remove $f else dpkg-statoverride --add root root 4755 $f fi done Additionaly, those lines in postinst are very troubling: " chown root.root /usr/lib/cgi-bin/openwebmail/openwebmail*.pl || true chmod 4755 /usr/lib/cgi-bin/openwebmail/openwebmail*.pl || true ", they should be removed as soon as possible. Under no circumstances should package install behave like this. -- System Information Debian Release: 3.0 Architecture: i386 Kernel: Linux forumakad 2.4.28-bsd25a #1 Thu Nov 18 11:54:59 CET 2004 i686 Locale: LANG=C, LC_CTYPE=C Versions of packages openwebmail depends on: ii apache1.3.26-0woody6 Versatile, high-performance HTTP s ii apache [httpd]1.3.26-0woody6 Versatile, high-performance HTTP s ii debconf 1.0.32 Debian configuration management sy ii libauthen-pam-perl0.12-2 This module provides a Perl interf ii libdbd-mysql-perl 1.2216-2 mySQL database interface for Perl ii libdbd-pg-perl1.01-3 a PostgreSQL interface for Perl 5 ii libmd5-perl 2.02-3 backwards-compatible wrapper for D ii libnet-ldap-perl 0.25-2 A Client interface to LDAP servers ii libtext-iconv-perl1.2-1 Convert between character sets in ii perl 5.6.1-8.8 Larry Wall's Practical Extraction ii perl-suid 5.6.1-8.8 Runs setuid Perl scripts. ii ucf 1.13 Update Configuration File: preserv ii wwwconfig-common 0.0.19 Debian web auto configuration. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#234526: munin-node: cpu graph draws very strange values on HT-enabled cpus
> > on dual-cpu ( actually HT-enabled p4 ), I get very strange results from > Sorry that I forgot about this bug. Does this still happen with the > 1.2.2 version? I have lots of servers with HT works just fine, see. I'll have to return to you on that in few days, I created my own cpu probe so I haven't seen if anything changed. -- Dariush Pietrzak, Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#661494: xfsrestore of xfsdump with inaccessible files fails. corrupt extent header
Package: xfsdump Version: 3.0.4 Severity: normal Problem exists in xfsdump version up to 3.0.6, it looks like this: time nice ionice -c 3 xfsdump - /device | xfsrestore - ./ xfsrestore: using file dump (drive_simple) strategy xfsrestore: version 3.0.6 (dump format 3.0) - Running single-threaded xfsdump: using file dump (drive_simple) strategy xfsdump: version 3.0.6 (dump format 3.0) - Running single-threaded xfsdump: WARNING: most recent level 0 dump was interrupted, but not resuming that dump since resume (-R) option not specified xfsdump: level 0 dump of pokurcz:/home xfsdump: dump date: Mon Feb 27 14:47:47 2012 xfsdump: session label: "" xfsrestore: searching media for dump xfsdump: ino map phase 1: constructing initial dump list xfsdump: ino map phase 2: skipping (no pruning necessary) xfsdump: ino map phase 3: skipping (only one dump stream) xfsdump: ino map construction complete xfsdump: estimated dump size: 41947280896 bytes xfsdump: creating dump session media file 0 (media 0, file 0) xfsdump: dumping ino map xfsdump: dumping directories xfsrestore: examining media file 0 xfsrestore: dump description: xfsrestore: mount point: /home xfsrestore: session time: Mon Feb 27 14:47:47 2012 xfsrestore: level: 0 xfsrestore: session label: "" xfsrestore: media label: "" xfsrestore: file system id: 9c620277-a63d-4a5b-bb82-64df562b6ddd xfsrestore: session id: 7bb62f13-dfd9-40ea-a0c4-43dcca7d7897 xfsrestore: media id: 1dd3dc69-4ef5-4fe5-8f5b-d8c4aeb7761f xfsrestore: searching media for directory dump xfsrestore: reading directories xfsdump: dumping non-directory files xfsrestore: 98210 directories and 645968 entries processed xfsrestore: directory post-processing xfsrestore: restoring non-directory files xfsdump: WARNING: could not open regular file ino 109308 mode 0x8180: Stale NFS file handle: not dumped xfsdump: WARNING: could not get list of non-root attributes for nondir ino 109308: Stale NFS file handle (116) xfsdump: WARNING: could not get list of root attributes for nondir ino 109308: Stale NFS file handle (116) xfsdump: WARNING: could not get list of secure attributes for nondir ino 109308: Stale NFS file handle (116) xfsrestore: WARNING: corrupt extent header xfsrestore: WARNING: unable to resync media file: some portion of dump will NOT be restored xfsrestore: restore complete: 27 seconds elapsed xfsrestore: Restore Status: SUCCESS xfsdump: ending media file xfsdump: media file size 198770688 bytes xfsdump: dump size (non-dir files) : 97237608 bytes xfsdump: NOTE: dump interrupted: 27 seconds elapsed: may resume later using -R option xfsdump: Dump Status: INTERRUPT after applying patch from http://www.spinics.net/lists/xfs/msg09707.html xfsdump works correctly ( produces files restorable with xfsrestore ) time nice ionice -c 3 xfsdump - /dev/s/home | xfsrestore - ./ xfsrestore: using file dump (drive_simple) strategy xfsrestore: version 3.0.6 (dump format 3.0) - Running single-threaded xfsdump: using file dump (drive_simple) strategy xfsdump: version 3.0.6 (dump format 3.0) - Running single-threaded xfsdump: WARNING: most recent level 0 dump was interrupted, but not resuming that dump since resume (-R) option not specified xfsdump: dump date: Mon Feb 27 15:41:58 2012 xfsdump: session id: e6ecdc0c-3173-4828-86a0-5e8df2a7f69c xfsdump: session label: "" xfsrestore: searching media for dump xfsdump: ino map phase 1: constructing initial dump list xfsdump: ino map phase 2: skipping (no pruning necessary) xfsdump: ino map phase 3: skipping (only one dump stream) xfsdump: ino map construction complete xfsdump: estimated dump size: 41947529920 bytes xfsdump: creating dump session media file 0 (media 0, file 0) xfsdump: dumping ino map xfsrestore: examining media file 0 xfsdump: dumping directories xfsrestore: dump description: xfsrestore: hostname: pokurcz xfsrestore: mount point: /home xfsrestore: session time: Mon Feb 27 15:41:58 2012 xfsrestore: level: 0 xfsrestore: session label: "" xfsrestore: media label: "" xfsrestore: file system id: 9c620277-a63d-4a5b-bb82-64df562b6ddd xfsrestore: session id: e6ecdc0c-3173-4828-86a0-5e8df2a7f69c xfsrestore: media id: 82113274-d47c-4504-a376-8107557f9068 xfsrestore: searching media for directory dump xfsrestore: reading directories xfsdump: dumping non-directory files xfsrestore: 98210 directories and 645992 entries processed xfsrestore: directory post-processing xfsrestore: restoring non-directory files xfsrestore: NOTE: ino 49280 gen 2421386218 not referenced: placing in orphanage xfsrestore: NOTE: ino 12297587 gen 2810501049 not referenced: placing in orphanage xfsrestore: NOTE: ino 13327059 gen 2649609090 not referenced: placing in orphanage xfsrestore: NOTE: ino 13328402 gen 3735435282 not referenced: placing in orphanage xfsrestore: NOTE: ino 13328403 gen 3735435348 not referenced: placing in orphanage xfsrestore: NOTE: ino 13328406 gen 3735435291 not referenced: placing in orphanage xfsrestore: NOTE: ino 36898991 gen 821504621 not refe
Bug#460310: vzctl: upgrade from older versions looses --name settings
> for n in $(find /etc/vz/names -maxdepth 1 -name "*.conf"); do > VEID= > m=`echo $n|sed -e "s/.conf//"` > . $n > echo $VEID > if [ -n "$VEID" ] ; then > ln -s /etc/vz/conf/$VEID.conf $m > fi > rm -f $n > done > > The /etc/vz/names/.conf file should be removed or? I would venture a guess that removing is fine, there shouldn't be any valuable data in that conf, only VEID number (which is preserved as symlink, so there is no risk of data loss). > Is there a good way to detect the change of config scheme? I'm only aware of testing using existance of *.conf settings in names/. Sorry for the long delay, I mistakenly marked this conversation as finished. -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 Total Existance Failure -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#495778: dovecot-pop3d: auth_debug_passwords = yes logs passwords even when they are correct.
Package: dovecot-pop3d Version: 1.0.rc15-2etch4 Severity: minor auth_debug_password is described as such: # In case of password mismatches, log the passwords and used scheme so the # problem can be debugged. Requires auth_debug=yes to be set. This means that only bad/erroneous authentications should be logged, however when auth_debug_password is set to 'yes', all conversations are logged -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18.bsd40x Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Versions of packages dovecot-pop3d depends on: ii dovecot-common 1.0.rc15-2etch4 secure mail server that supports m ii libc6 2.3.6.ds1-13etch7 GNU C Library: Shared libraries ii libssl0.9.80.9.8c-4etch3 SSL shared libraries dovecot-pop3d recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#460310: vzctl: upgrade from older versions looses --name settings
Hello, > Exactly what did you need to do to solve the problem? cd /etc/vz/names && for n in *; do echo $n; m=`echo $n|sed -e "s/.conf//"`; v=`cat $n|sed -e "s/VEID=.//"|sed -e "s/.$//"`; echo $m;ln -s /etc/vz/conf/$v.conf $m; done did the trick for me, I guess something logically equivalent wouldn't hurt much when old-style config is detected. -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 Total Existance Failure -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#354186: ignores hwrngdevice settings in /etc/default/.., typo in /etc/init.d/rng-tools
Package: rng-tools
Version: 2-unofficial-mt.10-1
Severity: normal
Tags: patch
This should fix it:
26c26
< [ -c "${HWRNGDEVICE}" ] && return 0
---
> [ -c "${HRNGDEVICE}" ] && return 0
-- System Information:
Debian Release: 3.1
Architecture: i386 (i686)
Kernel: Linux 2.4.32-bsd32e
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Versions of packages rng-tools depends on:
ii libc6 2.3.2.ds1-22 GNU C Library: Shared libraries an
ii makedev 2.3.1-77 creates device files in /dev
-- no debconf information
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#510188: [Pkg-samba-maint] Bug#510188: samba: Failed to set uid privileges to (-1, 2589) now set to (0, 0) in smbd log
> > on active production server it happens fairly often (~50 times a day), it > > started after upgrade. > > After what upgrade? Etch's samba hasn't been updated since > months. Maybe you were upgrading a sarge server to etch? My etch servers has been using sarge's samba, because of this bug: " - Erroneous permissions checks after 3.0.10 -> 3.0.14a (upstream #2591). Closes: #307626 " > > I'm fairly lost as to how should I proceed with trying to debug this, so > > any suggestions are welcome > > > Can you send back the output of "testparm /etc/samba/smb.conf" ? [global] workgroup = AD realm = MYAD.PL server string = Server %v interfaces = 192.168.10.10/255.255.255.0 security = ADS password server = ad1.myad.pl ad2.myad.pl log file = /var/log/samba/log.%m max log size = 50 debug uid = Yes socket options = TCP_NODELAY SO_SNDBUF=8192 SO_RCVBUF=8192 load printers = No dns proxy = No wins server = 192.168.10.12, 192.168.10.13 hosts allow = 127.0.0.1, 192.168.101.0/24, 192.168.100.0/24, 172.17.0.0/22 max connections = 21 [ebic_samba] comment = Samba.Test Directory path = /fs/samba/test valid users = user1, user2, @GROUP1, @GROUP2, user3, user4, user5, user6, @GROUP3 force group = GROUP2 read only = No force create mode = 0664 vfs objects = audit > (better than sending the raw smb.cofn file as this will "hide" options > that have the default values I'm starting to think that it might be client-dependant, as I wasn't able to cause the problem from any linux client, only from Windows 2000 -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 Total Existance Failure -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
