Bharat's profile - activity

Another possible solution.

I had this same issue. The internet is accessible via a proxy server and I had done many updates and OS upgrades without any issue. Suddenly my yum started throwing the HTTP 401 error. I finally figured out that I had to change proxy=http://<IP> in the /etc/yum.conf file to proxy=https://<IP>.

There was some changes in our proxy server settings that I was unaware of.

So this might also be one of the solutions. In general, if you face this error then try these (just a summary of different sources):

• run yum clean all
• add this line ipresolve=4 in /etc/yum.conf
• Check the proxy settings and set proxy address to https instead of http

So, are the other two OS working properly ? Does this problem recur?

is it a dualboot install ?

the question is not readable... please improve formatting...

I did try that but it didnt work. Perhaps because of lack of space.

I had updated fedora 17 to fedora 19 (kernel 3.10.11-200) on my laptop using fedup a ~2-3 weeks back. Everything was fine until yesterday night. All of a sudden while shutting down I got some recurring kernel error message. I couldn't note it down and I powered off the system forcibly. The next morning when I started the system the login screen fails to start. I am just able to see kernel messages. I get this message:

failed to start wait for plymouth boot screen to quit

When I start terminal with Ctrl+Alt+F2 i get this message when I log in:

login: write lastlog failed: No space left on device

yum updatedoesn't work and fails with this message:

error: db5 error(11) from dbenv->open: Resource temporarily unavailable
error: cannot open Packages index using db5 - Resource temporarily unavailable (11)
error: cannot open Packages in /var/lib/rpm
CRITICAL:yum.main:

Error: rpmdb open failed

I have around 50 GB for the root partition. Can't understand how can all this space run out. The disk is LVM partitioned. Is there any way to rescue the system without losing files on /home.

Thanks for the suggestions. I was too impatient then and I just reinstalled fedora. I should have checked what is taking up that much of space. Quite probable that it must have been something in /var/log/messages because during the last shutdown there was a torrent of messages.

superuser is a stackexchange site and it is quite easy to open a site there. However, it is not the basic structure of the site that is the issue. Issue is the basic rules (which is different is different SE sites), which can be changed by the mods.

superuser is a part of stackexchange. So it is quite easy to form a new SE. However, the issues that I point out are site-specific (even in SE sites) and can be changed by mods.

remove the space between http: and // .. i had to do it because i was not able to post it

Windows installation resets the MBR (Master Boot Record) so you wont be able to see your linux partitions. That is why the usual practice is to install windows and then linux. But it is not a big problem. You will have to reinstall grub (the linux boot loader) to the MBR using a live CD. Check the page mentioned below. It is quite informative.

http: //www.av8n.com/computer/htm/grub-reinstall.htm

I think Fedora-19 has quite some bugs (Fedora 17 was quite good). This time it is the XFCE panel. Though not a serious issue, the panel doesn't show all open windows (which i can see using alt+tab). My desktop has been set to show only minimized windows and I can see all of them there but not in the panel.

Checked panel preferences, but everything seems to be fine. Is it a bug ?

Get the sourcecode from www.rpm.org and compile it.

even if i codify the link address, I am not able to post it. I tried it. Yes I agree that killing wont solve the issue. As i said it is a temporary bruteforce solution: to allow yum upgrade at that instance.

Well. As Eddie says, it depends on how you partitioned your disk(s). If you have a separate /home partition then during the fedora installation, select manual partitioning and just repartition / and /boot while retaining /home from the previous partition. You might have to specify that you need to load the old /home partition. it will have a device number like sdaX or hdaX which you want to now load with label /home. you can do it by modifying /etc/fstab

Usually the ubuntu deault installation settings doesnt make a separate /home partition and in that case you wont be able to retain your /home contents in the new installation

There is another post that answers your question: see #10929 That post talks about how to disable PackageKit by editing the preferences.

There is another brute force way: Kill packagekit : Find PID by ps ax|grep packagekit and then kill <PID>

there are some strict karma requirements. first of all there there is no upvoting for a decent question. Most times people solve their own problems but cannot accept their answers because you need >20 karma. You cannot post links to substantiate the answer because it again requires karma. You cannot post comments as well.

What I dont like here is that badges (that are pretty useless and is just a kiddish gimmick) are easier to get than karma.

So it is natural to prefer superuser. [My personal experience: I posted same questions here and in superuser and it is the latter where I got responses. In here I just get some stupid badge]

updating initrd with dracut --force solved the problem.

Due to some problems, I had to reinstall fedora. I currently have fedora-19 with kernel version 3.11.4-201.

I haven't installed NVIDIA drivers and am using the default nouveau drivers for graphics. Bumblebee and bbswitch got installed properly but bbswitch fails to turn off discrete graphics. I did tee /proc/acpi/bbswitch<<<OFF but when I check the status it reports it on. dmesg|grep bbswitch gives this:

[   25.192983] bbswitch: module verification failed: signature and/or required key missing - tainting kernel
[   25.193158] bbswitch: version 0.7
[   25.193165] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[   25.193172] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[   25.193316] bbswitch: detected an Optimus _DSM function
[   25.193325] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
[ 1383.773601] bbswitch: device 0000:01:00.0 is in use by driver 'nouveau', refusing OFF

Bumblebee-daemon is running but optirun glxgears -info gives this:

[ 2441.794492] [ERROR]Cannot access secondary GPU - error: [XORG] (EE) Failed to load module "mouse" (module does not exist, 0)

[ 2441.794545] [ERROR]Aborting because fallback start is disabled.

I created:

• bbswitch.conf in /etc/modprobe.d/ with bbswitch load_state=0 in it.
• bbswitch.conf in /etc/modules-load.d/ with bbswitch in it.

• nouveau.conf in /etc/modprobe.d/ with

  blacklist nouveau

  blacklist nvidia

renamed nouveau.conf to blacklist.conf but no difference.

How do I switch off the discrete graphics card ? It happened quite smoothly in my previous fedora-17 which I fedupped to fedora-19. Isn't bumblebeed supposed to turn it off without manual intervention? How do I know if bbswitch is compatible with the kernel?

EDIT

During one boot the graphics was turned off but it turned on again when i tried optirun glxgears -info (which btw gave the same error). I reinstalled bbswitch and bumblebee but it doesnt solve the issue.