- Arch-general - lists.archlinux.org

Re: [arch-general] amd x86_64 crash: fuse init .. end_request: I/O error, dev fd0, sector 0?
by Andrej Gelenberg 27 Jul '10

27 Jul '10

Hi, try to get backtraces with sysrq-keys (w or l) or with kernel crash dump. Regards, Andrej On 07/26/2010 11:53 PM, David C. Rankin wrote: > On 07/26/2010 01:54 PM, Andrej Gelenberg wrote: >> Hi, >> >> On 07/26/2010 03:59 PM, David C. Rankin wrote: >>> Guys, >>> >>> Here is a possible bug regarding the fuse filesystem. The best I can >>> determine this occurred during an rsync request from the x86_64 box to a >>> laptop to retrieve files under /home/david. The box hardlocked. The >>> error messages just before the restart were: >>> >>> Jul 26 02:48:19 archangel kernel: fuse init (API version 7.13) >>> Jul 26 02:48:19 archangel kernel: end_request: I/O error, dev fd0, >>> sector 0 >>> Jul 26 02:48:19 archangel kernel: end_request: I/O error, dev fd0, >>> sector 0 >> >> I can't see, why it should be fuse bug. Have you mounted something >> with fuse? >> >> How run you the rsync? Maybe you try to sync /dev, and it cause the >> problems? >> >> Regards, >> Andrej >> > > That's what was strange to me as well. I just run rsync in the normal way: > > rsync -uav --ignore-existing otherbox:/the/dir/I/want ~/dir/here > > I'll keep an eye on it. It would have been something funky with the > power at that time. Even with a good ups, lightning is still king... >

1 0

[arch-general] amd x86_64 crash: fuse init .. end_request: I/O error, dev fd0, sector 0?
by David C. Rankin 26 Jul '10

26 Jul '10

Guys, Here is a possible bug regarding the fuse filesystem. The best I can determine this occurred during an rsync request from the x86_64 box to a laptop to retrieve files under /home/david. The box hardlocked. The error messages just before the restart were: Jul 26 02:48:19 archangel kernel: fuse init (API version 7.13) Jul 26 02:48:19 archangel kernel: end_request: I/O error, dev fd0, sector 0 Jul 26 02:48:19 archangel kernel: end_request: I/O error, dev fd0, sector 0 What stands out to me is 'fd0'. Why in the world would there be an I/O error to the floppy associated with a rsync request that had nothing to do with the floppy? The floppy on that box probably hasn't been used in over a year since the last bios upgrade and there is no floppy in the drive. This occurred with the following packages installed: kernel26 2.6.34.1-1 fuse 2.8.4-2 Has anybody seen anything similar? What is a good way to troubleshoot other than repeatedly doing rsync transfers and hoping to catch something? -- David C. Rankin, J.D.,P.E. Rankin Law Firm, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 Telephone: (936) 715-9333 Facsimile: (936) 715-9339 www.rankinlawfirm.com

2 1

[arch-general] Chroot 32bit under 64bit; Problems with kernel package
by Jordan Windsor 25 Jul '10

25 Jul '10

I'm following the guide available at this URL: http://wiki.archlinux.org/index.php/Arch64_Install_bundled_32bit_system Currently I'm stuck on this step: [root@archpc ~]# pacman --root /opt/arch32 --cachedir /opt/arch32/var/cache/pacman/pkg --config /opt/arch32/pacman.conf -S base base-devel error: failed to prepare transaction (package architecture is not valid) :: package kernel26-i915-2.6.34.1-1-x86_64 does not have a valid architecture I'm wondering how I should go about getting this chroot working. I'll be using the chroot to compile wine, I was unable to get it working with gcc-multilib. Thanks.

6 10

[arch-general] New to Arch with problems with AUR
by Mario Daniel Carugno 25 Jul '10

25 Jul '10

Hi list ! First i present myself. My name is Mario and i was reading this list for a few weeks. I'm coming from Debian, and i have to say that Arch was the first distro that replaced it in my computer. Arch rocks, it gives a really updated and stable kde desktop. Yes, it's a little hard to configure, but one can learn with it. And the AUR system was one of the main reasons for me to change. I was trying to install some programs from AUR, and it worked in some cases. But in many other cases, the programs do not compile. And i talk about programs with so much votes. For instance, i couldn't compile shaman. Actually i can't compile aqpm or libaqpm, wich is a dependency of shaman, and it isn't in the pacman repositories. So, in general, how must i face these problems ? Do i have to post errors in the AUR pages ? Is it always so common to have problems in building AUR packages ? By the way, i will stay with Arch. I like it's descentralized package philosophy. Thank you

5 4

[arch-general] X11 forwarding: X Error: BadAccess (attempt to access private resource denied) 10
by David C. Rankin 24 Jul '10

24 Jul '10

Guys, I'm trying to use x11 forwarding and from my arch laptop and I can't run X apps from my arch server. I know this has to do with the MIT-MAGIC-COOKIE-1 or something similar, but I can't remember what we did to fix this last time. What's the trick to get it going?? ICEauthority? pam entry? -- David C. Rankin, J.D.,P.E. Rankin Law Firm, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 Telephone: (936) 715-9333 Facsimile: (936) 715-9339 www.rankinlawfirm.com

5 8

[arch-general] [PATCH 1/4] Add skeleton for hooks support
by Victor Lowther 23 Jul '10

23 Jul '10

This begins work on adding some sort of hook infrastructure to pacman. There has been much discussion and little code that I have seen, so I figured I would go ahead an implement a very basic skeleton for transaction hooks support. --- lib/libalpm/Makefile.am | 1 + lib/libalpm/{sync.h => hooks.c} | 34 +++++++++++++++++++++++++--------- lib/libalpm/{add.h => hooks.h} | 19 ++++++++----------- 3 files changed, 34 insertions(+), 20 deletions(-) diff --git a/lib/libalpm/Makefile.am b/lib/libalpm/Makefile.am index 3473a73..b4fdadf 100644 --- a/lib/libalpm/Makefile.am +++ b/lib/libalpm/Makefile.am @@ -44,6 +44,7 @@ libalpm_la_SOURCES = \ sync.h sync.c \ trans.h trans.c \ util.h util.c \ + hooks.h hooks.c \ version.c libalpm_la_LDFLAGS = -no-undefined -version-info $(LIB_VERSION_INFO) diff --git a/lib/libalpm/sync.h b/lib/libalpm/hooks.c similarity index 65% copy from lib/libalpm/sync.h copy to lib/libalpm/hooks.c index 000a09c..9630c70 100644 --- a/lib/libalpm/sync.h +++ b/lib/libalpm/hooks.c @@ -1,9 +1,10 @@ /* - * sync.h + * hooks.c * * Copyright (c) 2006-2010 Pacman Development Team <pacman-dev(a)archlinux.org> * Copyright (c) 2002-2006 by Judd Vinet <jvinet(a)zeroflux.org> * Copyright (c) 2005 by Aurelien Foret <orelien(a)chez.com> + * Copyright (c) 2005 by Christian Hamar <krics(a)linuxforum.hu> * Copyright (c) 2005, 2006 by Miklos Vajna <vmiklos(a)frugalware.org> * * This program is free software; you can redistribute it and/or modify @@ -19,14 +20,29 @@ * You should have received a copy of the GNU General Public License * along with this program. If not, see <http://www.gnu.org/licenses/>. */ -#ifndef _ALPM_SYNC_H -#define _ALPM_SYNC_H -#include "alpm.h" - -int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync, alpm_list_t **data); -int _alpm_sync_commit(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t **data); +#include "config.h" -#endif /* _ALPM_SYNC_H */ +#include <stdlib.h> +#include <stdio.h> +#include <string.h> +#include <unistd.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/statvfs.h> +#include <errno.h> -/* vim: set ts=2 sw=2 noet: */ +/* libalpm */ +#include "hooks.h" +#include "trans.h" +#include "alpm_list.h" +#include "package.h" +#include "util.h" +#include "log.h" +#include "handle.h" +#include "add.h" +#include "remove.h" +#include "sync.h" +#include "alpm.h" +#include "deps.h" +#include "cache.h" diff --git a/lib/libalpm/add.h b/lib/libalpm/hooks.h similarity index 73% copy from lib/libalpm/add.h copy to lib/libalpm/hooks.h index eb37dc7..d60b1f4 100644 --- a/lib/libalpm/add.h +++ b/lib/libalpm/hooks.h @@ -1,8 +1,11 @@ /* - * add.h + * hooks.h * * Copyright (c) 2006-2010 Pacman Development Team <pacman-dev(a)archlinux.org> * Copyright (c) 2002-2006 by Judd Vinet <jvinet(a)zeroflux.org> + * Copyright (c) 2005 by Aurelien Foret <orelien(a)chez.com> + * Copyright (c) 2005 by Christian Hamar <krics(a)linuxforum.hu> + * Copyright (c) 2006 by Miklos Vajna <vmiklos(a)frugalware.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -17,15 +20,9 @@ * You should have received a copy of the GNU General Public License * along with this program. If not, see <http://www.gnu.org/licenses/>. */ -#ifndef _ALPM_ADD_H -#define _ALPM_ADD_H +#ifndef _ALPM_HOOKS_H +#define _ALPM_HOOKS_H -#include "db.h" -#include "alpm_list.h" -#include "trans.h" +#include "alpm.h" -int _alpm_upgrade_packages(pmtrans_t *trans, pmdb_t *db); - -#endif /* _ALPM_ADD_H */ - -/* vim: set ts=2 sw=2 noet: */ +#endif /*ALPM_HOOKS_H */ -- 1.7.1.1

2 6

[arch-general] A question about using tar... ( -p & --numeric-owner)
by Joe(theWordy)Philbrook 23 Jul '10

23 Jul '10

I'm a multi-linux-distro/multi-boot guy. I always have more than one configured linux distro on my grub menu. Whenever I'm considering radical changes to one of my working linux I like to back it up. Quite some time ago I learned to do this with tar. For example I've got both Arch Linux and Xubuntu on this desktop pc. I've decided to attempt using a network upgrade process to upgrade the Xubuntu from Karmic to the current Lucid release. So first I boot Arch, mount a storage drive on /mnt/other, & mount the Xubuntu partition on /mnt/xubuntu. Then cd to /mnt/xubuntu. (with a root shell of course) # tar -czf /mnt/other/xubuntu.tgz -p --numeric-owner . And of course if it's necessary I can restore the backup to an empty partition with: # tar -xzf /mnt/other/xubuntu.tgz -p --numeric-owner . But I only partially understand the -p and --numeric-owner options. I do know that -p is supposed to be default for root but I like to make sure... I think I understand that the reason somebody told me (quite some time ago) to use --numeric-owner is because (in this example) Arch may have different user id numbers than Xubuntu which could have bad results if it tried to apply permissions based on user names (especially if it involves a "system account")... What I don't know is if either of these options are necessary during the "tar -czf" operation, or if I only need to bother with them when/if I restore a file system with "tar -xzf" ??? I know it does work the way I expect it to when I use the options for both the "-czf" & the "-xzf" operations. And that I don't get any error messages from tar when I do it that way. But I don't know if using them for the "-czf" operation makes any difference or not. Would someone be so kind as to set me straight on this??? -- | ~^~ ~^~ | <?> <?> Joe (theWordy) Philbrook | ^ J(tWdy)P | \___/ <<jtwdyp(a)ttlc.net>>

1 0

[arch-general] Clarfication request regarding mkinitcpio code
by Dave Reisner 22 Jul '10

22 Jul '10

There's a few things in mkinitcpio I'd like to cleanup, but I need to have one issue clarified before I proceed. The header of mkinitcpio itself states that various parts need to run under Busybox's ash, and thus various standards need to be adhered to. However, the file itself sports a /bin/bash shebang, and makes mild use of parameter expansion, e.g. APPNAME=${0##*/} on line 44. I understand that hooks themselves need to run in a more POSIX environment, but would it be safe to use more Bash syntax in the code that generates the initcpio image? I would be sticking to a feature set compatible with 3.2. thanks, dave reisner

1 0

[arch-general] Hard lockups during boot
by Chris Bannister 22 Jul '10

22 Jul '10

Hi, I used to use a zen kernel but that stopped working which is an unrelated story so I switched back to the Arch kernel, but im getting hardlocks during boot. It happened twice just now, first time was after ATA got init'd but before the SCSI came up. The second time it got to the waiting for udev .. during userspace init. Heres the attached working dmesg log [1] in case its of anyhelp. I didnt want to post this to the bug tracker because of the lack of logs I can provide so ill put it here and if anyone has any ideas please shoot them over. Thanks! [1] http://zariel.pastey.net/138823

1 0

[arch-general] simple iptabels question
by aerospace1028＠hotmail.com 22 Jul '10

22 Jul '10

Hello, I have a question on iptables. On the arch wiki, the tutorial on a simple stateful firewall includes the creation of the following rule. # iptables -A INPUT -p tcp --syn -m state --state NEW -j OPEN-TCP Which, to my understanding, meant only pass new TCP streams with the syn flag (the initialize connection or handshake part 1 of 3) to the user defined chain. When I checked the output of iptables, I noticed the following rule. # iptables -L INPUT --line-numbers -v 7 OPEN-TCP tcp -- anywhere anywhere tcp flags:FIN,SYN,RST,ACK/SYN state NEW Why is iptables also using the FIN, RST and ACK/SYN flags? Did I not request SYN only like I thought the tutorial said? _________________________________________________________________ The New Busy is not the old busy. Search, chat and e-mail from your inbox. http://www.windowslive.com/campaign/thenewbusy?ocid=PID28326::T:WLMTAGL:ON:…

2 1