users June 2011

users@lists.fedoraproject.org

351 participants
514 discussions

14->15 upgrade comments
by Rich Emberson 01 Jun '11

01 Jun '11

Last weekend I upgraded (using preupgrade) both my laptop and development box from 14 to 15. Both are x86_64 machines. On the laptop, twice now while rebooting the system has crashed with a stacktrace on the screen. Both times it occurred while the blue, light- blue and white progress bars were making their way across the screen. I do not know howto capture the stacktrace. The machine freezes. Ctl-Alt-Del does not cause a reboot. The offswitch does not work. After trying both of the above and waiting a couple of minutes, the machine turns off. Afterwards, I attempt another reboot and it works. Must be some timing issue. (The first time it happened was during the reboot at the end of the preupgrade process - needless to say my stress levels spiked.) Secondly, on my development machine I have a 4TB drive that I always mounted manually; its not in the /etc/fstabs file. Well, while rebooting at the end on the preupgrade process, I got the following popup: Dirty File Systems The following file systems for your Linux system were not unmounted cleanly. Please boot your Linux installation, let the file systems be checked and shut down cleanly to upgrade. /dev/sde1 Clicking OK, I was booted into 14. I then ran fsck on the disk which took many hours with no errors. Then, had to re-run the preupgrade script (which was much faster the second time since things had already been downloaded) and the reboot took me to 15. You might consider letting folks know on the preupgrade webpage that all disks, not just those in fstabs, HAVE to have had fsck run recently (don't know what recent means in this case). Lastly, the fonts are, well, not what they were before the upgrade. Others have mentioned this and I will live with them until some future time. Other than those issues, great upgrade process. Many Thanks! Richard

4 3

btrfs advice?
by Michael Wiktowy 01 Jun '11

01 Jun '11

I made the leap to a btrfs root partition for my netbook with a fresh install of F15. Everything seems to work fine and I have done some fair amount of Googling for information and come across https://btrfs.wiki.kernel.org/index.php/Getting_started which offers some great (incomplete) info but nothing Fedora specific. I have some questions about how to move forward to take advantage of btrfs: 1) Is there any kind of integration of btrfs with the included apps (For example: automatic snapshots before yum updates allowing easy rollbacks, deja dup backing up a btrfs snapshot so that currently changing data doesn't affect the backup process, a kiosk mode that rolls back a home directory to a known state after logout, etc.) or is that the next step to take advantage of all the new bells and whistles and F15 is just a test-btrfs-as-a-ext4-replacement release? 2) I see (using 'mount') that the root is btrfs along with some (what appear to be) subvolumes for /tmp, /var/tmp and /home yet 'btrfs device scan' shows no information. Does 'btrfs device scan' only scan unmounted devices or is this a bug? 3) I have read at Phoronix that using the transparent compression offers a fair performance gain ( http://www.phoronix.com/scan.php?page=article&item=btrfs_space_cache&num=1 ) with mixed results when combining that with space cache. Has the space cache+compression degradation seen in some benchmarks been fixed in F15? Also, is using transparent compression simply a matter of adding the correct mount flag to /etc/fstab or is there a more complex conversion process to be followed that needs to be done offline? 4) Since this is a netbook with a modest Atom processor, would enabling transparent compression just load the CPU such that any performance is negated or cause significant battery drain? Thank you for any experience you have to offer. /Mike

4 5

RE: Routing not working
by Alan Gagne 01 Jun '11

01 Jun '11

> My next step is finding the digital thermometer we have around > here somewhere and measuring the temp on f14 compared to f15 :-). Would that be oral or otherwise? Seems otherwise might be appropriate given the level of fun this release has provided the users. Alan

2 1

F14 NetworkManager IPv6 behavior question
by Jurgen Kramer 01 Jun '11

01 Jun '11

I just enabled IPv6 on my DSL router and nicely got an non-tunneled IPv6 address from my provider. After enabling IPv6 in NM it tried getting a IPv6 address using DHCPv6. As I dont need DHCP for IPv6 I configured the connection in NM to 'automatic'. When I check with ifconfig I see that I actually have a working IPv6 address but NM itself does not list it. The connection information for IPv6 is empty. Checking with dmesg a see that NM (?) still issues the DHCPv6 command (which fails) while the interface already got a working address. NetworkManager[25794]: <info> dhclient started with pid 25823 Internet Systems Consortium DHCP Client 4.2.0-P2 Copyright 2004-2010 Internet Systems Consortium. All rights reserved. For info, please visit https://www.isc.org/software/dhcp/ Bound to *:546 Listening on Socket/eth0 Sending on Socket/eth0 PRC: Requesting information (INIT). XMT: Forming Info-Request, 0 ms elapsed. XMT: Info-Request on eth0, interval 970ms. XMT: Forming Info-Request, 970 ms elapsed. XMT: Info-Request on eth0, interval 2010ms. XMT: Forming Info-Request, 2980 ms elapsed. XMT: Info-Request on eth0, interval 4130ms. XMT: Forming Info-Request, 7110 ms elapsed. XMT: Info-Request on eth0, interval 8030ms. XMT: Forming Info-Request, 15150 ms elapsed. XMT: Info-Request on eth0, interval 15730ms. XMT: Forming Info-Request, 30900 ms elapsed. XMT: Info-Request on eth0, interval 32510ms. NetworkManager[25794]: <warn> (eth0): DHCPv6 request timed out. NetworkManager[25794]: <info> (eth0): canceled DHCP transaction, DHCP client pid 25823 Before I go to bugzilla, is this expected behavior ? Thanks, Jurgen

1 0

f15 kernel on f14?
by Alex 01 Jun '11

01 Jun '11

Hi all, Is it possible to use the fc15 kernel on fc14, and if so, how would you go about doing that? I could of course just download the SRPM and recompile it, but I'd like to be able to easily update the system as well. I don't doubt other RPMs will be required, but I can't upgrade this system just yet, but still want to benefit from the much later kernel. Are there too many dependency changes, such as module-init-tools, grub, initscripts, etc, that woud normally be tied to the kernel version to upgrade to the new one? Thanks, Alex

6 7

Repository Managers lunch at OR in Austin, Wed. June 8
by Laherty, Jennifer 01 Jun '11

01 Jun '11

Hello! Sara Fuchs at GA Tech and I are keen to meet with other repository managers attending the Open Repositories conference next week in Austin. If you're going to be there and have time during Wednesday's lunch to gather together to chat, we'd be delighted to meet you. I will wear a very orange t-shirt and should be easy to spot. The lunch runs from 12-1:30. We can start out in Salon C and if the desire is to move, we will. Managers running systems on any type of software are all welcome. Hope to see you there, Jennifer and Sara Jennifer Laherty Digital Publishing Librarian, IUScholarWorks Indiana University Bloomington Libraries Wells Library E159 Bloomington, IN 47405 812-855-5609

1 0

Logwatch question on Fedora 15
by Mark C. Allman 01 Jun '11

01 Jun '11

On earlier Fedora releases if I typed "logwatch --logfile maillog" then logwatch would only look at the log file group "maillog." I just updated to F15 over the weekend and now the switch doesn't seem to work. On F15 I get a full logwatch output if I type the above command. Can someone check this and reply if they're also seeing this? I googled around and didn't find anything. From the man page it still looks like the "--logfile" option should behave as it has in the past. $ uname -a Linux fornax 2.6.38.6-27.fc15.x86_64 #1 SMP Sun May 15 17:23:28 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux Thanks, -- Mark C. Allman, PMP, CSM Allman Professional Consulting, Inc. First Vice-President, Ocean State PMI www.allmanpc.com, 617-947-4263

1 0

VirtualBox bug ?
by Luc MAIGNAN 01 Jun '11

01 Jun '11

Hi, I use VirtualBox 4.0.8 to manage linux virtual machines. VirtualBox seems to hang up one virtual machine randomly. Has anyone have the same behavior ? Is there a way to have a workaround ? BR

2 1

Re: /var/log/messages question
by R. G. Newbury 01 Jun '11

01 Jun '11

> On Mon, 2011-05-30 at 18:36 -0400, Tom Horsley wrote: >> > Just for curiosity, why has no one ever arranged to >> > log something like "Hey! I'm booting the system again!" >> > as the very first line in that goes to /var/log/messages >> > when the system is booting? > I used to do something like that back when I was still using another OS. > I modified the bootup script to add my own entry at the beginning. You > could do the same, I don't think the script gets changed by updates that > often. It is far easier to add: echo "Completed boot at `date`" to the bottom of /etc/rc.d/rc.local Then it does not matter whether the system is using the old method or systemd and whether it is writing to 'messages' or boot.log. R. Geoffrey Newbury

7 10

Shared to other computers
by Jack Howarth 01 Jun '11

01 Jun '11

Has anyone managed to get the "Shared to other computers" option described in... http://www.techerator.com/2010/01/how-to-setup-internet-connection-sharing-… to work under Fedora 14 or 15 when both network interfaces are ethernet devices? I have an Asus A8N-SLI Premium motherboard with eth1 attached to the external ethernet and eth0 configured in the Network Connections dialog to use the "Shared to other computers" method. When I attach a Fedora 15 i386 laptop to eth0 via an ethernet patch cable, under Fedora 14 the connection was eventually made but the dns didn't work. Under Fedora 15, the laptop shows a red X for the network connection. The documentation on this method of internet sharing with two ethernet cards is very sparse and very unclear on whether additional dnsmasq or dhcp support is required to be configured. Jack

1 0

Jump to page: