Linux - ServerThis forum is for the discussion of Linux Software used in a server related context.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
We have a Red Hat 9 box that will no longer boot. The error being
Code:
init: relocation error: /lib/tls/libc.so.6: symbol _dl_starting_up, version GLIBC_PRIVATE not defined in file ld-linux.so.2 with link time reference
Kernal Panic: Attempted to kill init!
Something like this should not just occur "out of the blue" by itself. Has anyone recently (prior to this problem) tried upgrading (or downgrading) any packages installed in the machine (such as GLIBC?)? Have you otherwise changed any software or binaries that are installed?
Based on a very quick Google search for similar errors, it's difficult to determine exactly why this is "suddenly" happening, other than something has happened to the glibc components, so any further information you can provide might be very helpful.
Something like this should not just occur "out of the blue" by itself. Has anyone recently (prior to this problem) tried upgrading (or downgrading) any packages installed in the machine (such as GLIBC?)? Have you otherwise changed any software or binaries that are installed?
Based on a very quick Google search for similar errors, it's difficult to determine exactly why this is "suddenly" happening, other than something has happened to the glibc components, so any further information you can provide might be very helpful.
Thanks,
I believe one of our developers may have tried to updated PHP5.1 to PHP5.2 on the box. I remember him saying something about trying to install rpm, apt-get or yum. I was way too busy to overlook or assist him (not that my linux skills are great anyway).
Hmm, well, it would be less difficult of a fix, if we knew exactly what he did to the machine -- and what packages were affected by what he did. Perhaps talking to this developer fellow would be the best first step in determining what happened; after that, I suspect you'll be wanting to repair (reinstall some stuff) using the original installation media if possible - that would be the usual way of repairing a damaged system.
BUT: I personally am not experienced in upgrading RedHat systems, so I'm not sure how much trouble it would be to fix this if your RH license is expired, since (to my understanding - I could be off-base here) RH tends to not allow (or at least makes it difficult to do) software upgrades/installations when it is not under an active service contract.
If you do have a service contract with RH, contact them for advice. If not, then hopefully another LQ member will be familiar with that version to offer good advice.
By the way, RH 9 appears to be extremely old - like circa 2003? - so you may want to consider updating to something newer, or perhaps considering CentOS as a free alternative.
In any event, either talk to that developer guy to see if he has any insight, or wait for an experienced RH user here to offer some tips about what you might do first to fix this.
I have decided to blow it away and put Debian on the system. However I need to backup the required content on the server before hand. I have booted a Ubuntu Live CD on the server and can see the local hard disks. However I do not have permissions to copy files from them to a network share.
I'm don't want to advise about the "Network Share" portion, because that depends on what you've got set up, and I will presume you know what you're doing there.
As for the permissions issue, try doing what you're doing (the copying) as root (with the ubuntu live CD, you'd want to use `sudo` and start a nautilus or a terminal from which to work as root). If that doesn't work, you *might* be able to CHROOT into the messed up machine's root partition, and copy stuff out from there; this may or may not work, depending how messed up the machine is.
If neither of these seem to work, or you still can't get it, let us know and we'll try to come up with another idea.
I'm don't want to advise about the "Network Share" portion, because that depends on what you've got set up, and I will presume you know what you're doing there.
I have setup a samba share on another server and have connected to it from the Ubuntu live cd fine. It's wide open no restrictions.
Quote:
Originally Posted by GrapefruiTgirl
As for the permissions issue, try doing what you're doing (the copying) as root (with the ubuntu live CD, you'd want to use `sudo` and start a nautilus or a terminal from which to work as root). If that doesn't work, you *might* be able to CHROOT into the messed up machine's root partition, and copy stuff out from there; this may or may not work, depending how messed up the machine is.
The problem I have I guess is that RHEL was used to create all the files on it's local disk using root and a password. I'm connecting to them using the Ubuntu live cd and not giving that root password that it requires for access.
Edit: Put it this way, if a linux boxes OS gave up and it was irrepairable/unbootable natively, how would you get the data from it?
Edit: Put it this way, if a linux boxes OS gave up and it was irrepairable/unbootable natively, how would you get the data from it?
I would do what you're doing - use a live CD or another installed OS on the same machine, and as root, I would mount the damaged partition, and the backup partiton, and then proceed to copy what I wanted. It has always worked for me and you should not need the password for the damaged machine in order to mount that damaged partition.
I would do what you're doing - use a live CD or another installed OS on the same machine, and as root, I would mount the damaged partition, and the backup partiton, and then proceed to copy what I wanted. It has always worked for me and you should not need the password for the damaged machine in order to mount that damaged partition.
I can mount the damaged partition. Infact it mounts itstlf when I boot from the live CD. I can browse the internal disks, however when I attempt to copy from them to the network share I get permissions errors. The errors are pertaining to the local disks and not the remote share. I can create files, move/delete files on the remote share fine.
I have even tried
Code:
sudo passwd root
Whilst in the livecd session to change the root password to the same as the root password on the system. Still no change.
The fact that the partition is already mounted at boot of the live CD, *might* be pointing to the root of the problem. Please try, as root, umounting and re-mounting the partition in question, and then see if you have access.
Basically:
1 - Boot live CD
2 - Become root (`sudo su` or `sudo nautilus` etc..)
3 - umount the partition.
4 - remount the partition.
5 - try to copy.
There might be an issue involving PAM (Pluggable Access Module) here, which is a security measure. If this is the case, I am unable to advise further since my system does not use PAM. Someone else might be able to suggest what to do if that starts appearing to be the case.
Jeez.. I'm sorry all this hasn't helped; I really do not know what else to suggest. This routine has worked for me many times on my home machine(s). There must be either something painfully obvious we're both missing, or something about the RH installation that I/we don't know, which is doing this.
Did you happen to try chroot'ing into the RH box? If so, what was the result?
(In case you're unfamiliar, you would:
- become root on the Ubuntu CD.
- get a console as root
- mount the RH partition
- as root, type: chroot /some/mount/point #wherever you mounted the partition.
- now, you're effectively running "inside" the RH system, not the Ubuntu one.
- inside the chroot, mount (or connect) the location where you will copy to.
- begin copying.
- if this works, umount the destination when done & type "exit" to exit chroot and return to Ubuntu.)
And if this doesn't work, as mentioned, I'm at a loss unfortunately. I'm (fairly) sure someone else will have a clue what the problem is here, but we'll have to wait for them to appear.
Last edited by GrapefruiTgirl; 07-23-2010 at 10:54 AM.
Reason: added exit details
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.