Which Of This Files In /boot Is Safe To Be Deleted
Dec 4, 2007
After installing Kernels seems /boot is boing to be full soon. Which files can be safely deleted ? Mean lets say to keep the latest Kernel and previous version:
I have WHM 11.1.0 cPanel 11.2.1-C11635 FEDORA 4 i686 - WHM X v3.1.0
My /var partition is over 64% full. 2 directories in there show a lot of space being used.
One of these is /var/lib/mysql du -h --max-depth=1 # shows user accounts on the server with their database names. Are these just logs, or are they the actual databases? (Probably dumb newbie question)
The other big one is /var/log/munin Is there anything there I can delete?
My /var partition is getting full and most of the problem seems to be with the files in /var/cache/logwatch/ using up all the space. Can those be deleted?
The "/etc/php.ini" file seems reasonable and the session file path seems correct: session.save_path = "/var/lib/php/session"
But the session files never get deleted automatically. The session folder fills up quickly and I have been deleting session files manually while trying to resolve the problem. How can I verify that the default Plesk cron jobs are set up and running properly?
When you delete a site backup from its "Backup Manager" Panel, it is removed and no longer displayed in the Panel. However, I cannot tell if this action actually does anything with the real site backup files in "/var/lib/psa/dumps". Does this action merely remove it from PSA's database but not touch any actual files? If this is true, then how are site backup files supposed to be managed if this action doesn't actually delete them?
I've had some recurring problems with my host, VPSLink. For some reason startup scripts and init files keep on being deleted and I can't reach my site. I am indeed a bit of an amateur at maintaining a site but I haven't messed with these files or deleted them. What could be going on? An attack from someone on the web? Or some kind of VPSLink related problem?
This is part of the latest reply I got from VPSLink:We have managed to get your VPS back online. It appears some Ubuntu package changed the way networking is started/shut down which removes the /var/run/network directory completely. This directory contains the 'ifstate' file, which OpenVZ uses to set up the network interfaces.
The site is indeed up and running again so I'm not desperate, but I would very much like to understand what's going on here.
I've supposedly set APF firewall to start at boot time, by doing something like:
chkconfig --levels 2345 apf on
However i have my reservations to weather it is actually starting, its set to block port 80, after boot if i try and access it, the connection will get refused straight away, however if i go and manually start APF then try and access again, it will take a while, like its ignoring the connection attempt (which is good).
I'm trying to put together a new 1U rackmount server that I'm intending to colocate (my first server of my own). Here are the details:
11-152-087 CHASSIS SUPERMICRO|CSE-512L-260B R 13-182-120 SERVER_MB SUPERMICRO|MBD-PDSML-LN2+ 22-144-417 HD 250G|WD 7K 8M SATA2 WD2500JS 19-115-029 CPU INTEL|C2D E6750 2.66G 65N 4M R 20-134-337 MEM 1Gx2|KST KVR667D2E5K2/2G R
The stock CPU fan is obviously too big to sit in a 1U rackmount chassis, but I've got the top off and was just trying to boot up with the OEM fan to make certain everything worked (forgot to order a low-profile fan and heatsink--they're on the way).
Well, I put everything together, and it won't boot. Upon power up, there is nothing sent to the VGA port, and after a second or two, there is a hi-low alarm sound that does not stop that seems to imply the system is overheating (the MB and chassis manuals don't mention this alarm anywhere, but the read "temperature" light on the front is red). The CPU fan runs as well as the case fan, and even if I hold a huge high-velocity home fan over it, it still won't send any signal to the VGA monitor or POST.
I've checked that the power connections are good, the fans are seated properly, the memory contact is good, etc., and all I get is this hi-low chime every time I try to boot up. I think it's either a bad MB or CPU, since I can't imagine that the CPU is overheating with the heat sink and CPU fan.
Perhaps I've screwed something up, but the next issue I have is trying to figure out what could be causing the problem so I know what to return or replace. Is it the CPU? MB? RAM? Power unit? How can I test any of these without buying another set and possibly frying those?
I figured maybe some of you might have some ideas. I've built several computers over the years, but this one has me completely stymied.
This is probably the wrong location for this thread but I am unsure which section it is best located at.
I have 2 hard drives - 1 has XP pro installed (master drive) on it and the other is blank and set to slave. On the blank one i intedn to install Red Hat linux.
How is the best way to create a dual boot with these machines if each hard drive has one O/S dedicated to it.
I'm not sure exactly how to phrase the question. But, I'm researching how to PXE boot a server without having a DHCP/PXE server in each vlan.
Scenario: Datacenter with dozens of servers. 1 VLAN per server. Cisco switches and routers. Each server has a serial console available for remote management (OS and BIOS are configured for serial console). If an admin wants to re-install OS, they should be able to reboot the server and tell the BIOS to initiate a PXE boot request. A central install server is available to provide the DHCP and PXE boot images.
Has anyone tried this? I have been reading about the 'ip helper-address' for Cisco to relay DHCP requests. Interested in hearing about real-world setups. Or is there a better way to accomplish remote OS installs?
We cannot figure out why our dedicated server will not boot to the correct kernel. I've removed all other options from grub.conf but it's still booting to the default CentOS setup.
grub.conf:
Code: # grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You do not have a /boot partition. This means that # all kernel and initrd paths are relative to /, eg. # root (hd0,0) # kernel /boot/vmlinuz-version ro root=/dev/mapper/ddf1_4c53492020202020808627c300000000378494a900000a28p1 # initrd /boot/initrd-version.img #boot=/dev/mapper/ddf1_4c53492020202020808627c300000000378494a900000a28 default=0 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.18-028stab062.3) root (hd0,0) kernel /boot/vmlinux-2.6.18-028stab062.3 ro root=LABEL=/ initrd /boot/initrd-2.6.18-028stab062.3.img
I've got Centos 4 and I'm wondering what's the best way to cleanup my /boot partition?
Tried to do a yum update tonight and it included kernel, amongst other updates that belonged there so it stopped. I've googled around for commands to run and whatnot, but no go... or I just can't find it... if I had to clean it up I have an idea already about what to do, but I want to ask for advice first to see if there's an easier way.
on a RHE 3 system I installed a new Kernel. I did update lilo.conf and grub.conf but they are still booting an old kernel: 2.4.21-27.ELsmp. Please have a look at my files below and if you have any idea why please let me know.
Code: default=0 timeout=10 splashimage=(hd0,0)/grub/splash.xpm.gz title Red Hat Enterprise Linux ES (2.4.21-53.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-53.ELsmp ro root=/dev/hda3 initrd /initrd-2.4.21-53.ELsmp.img title Red Hat Enterprise Linux ES (2.4.21-47.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-47.ELsmp ro root=/dev/hda3 initrd /initrd-2.4.21-47.ELsmp.img title Red Hat Enterprise Linux ES (2.4.21-47.EL) root (hd0,0) kernel /vmlinuz-2.4.21-47.EL ro root=/dev/hda3 initrd /initrd-2.4.21-47.EL.img title Red Hat Enterprise Linux ES (2.4.21-27.ELsmp) root (hd0,0) kernel /vmlinuz-2.4.21-27.ELsmp ro root=LABEL=/ initrd /initrd-2.4.21-27.ELsmp.img title Red Hat Enterprise Linux ES-up (2.4.21-27.EL) root (hd0,0) kernel /vmlinuz-2.4.21-27.EL ro root=LABEL=/ initrd /initrd-2.4.21-27.EL.img
am running CentOs with whm/cpanel and am whenever i reboot my server all sites that required mysql show erros ( can't connect ) it resolves by manually in the web host manager restarting mysql.
I have an old server with only IDE and SCSI connectors on the mobo.
However, I have only SATA drives and no IDE available. I consider to buy an entry level PCI SATA adapter card. What do you think ? Can I install and boot Linux directly from it?
Starting Tomcat server Output from "/sbin/rc3.d/S998Tomcat start": ---------------------------- Using CATALINA_BASE: /opt/mediation/Tomcat5.5 Using CATALINA_HOME: /opt/mediation/Tomcat5.5 Using CATALINA_TMPDIR: /opt/mediation/Tomcat5.5/temp Using JRE_HOME: /opt/mediation/java/jre15_15002 /opt/mediation/Tomcat5.5/bin/catalina.sh[233]: touch: not found. ----------------------------
If Tomcat is manually stopped and restarted, it accepts connections.
I have set up Windows Small Business Server 2003 and was testing it over the past week. Everything has been going well and yesterday it requested to install an update (not to sure what it was now) and soon after the update was installed a restart was required.
After the server shutdown and tried to restart the system will not restart.
The machine powers up and then I am faced with a blank black screen (no load bar nothing). The monitor led light flashes from green to orange then to a static green light after the Dell splash screen.
I can load up into safe mode although I do not know how to work out what is wrong. I am guessing the update has caused the issue, possibly a boot sector needs repairing. There are no new additions to control panel -> install/remove programs which i believe is where all updates can be uninstalled from.
I backed a system up with rsync, and just used rsync to copy it into a new machine... I just backed up / to my home directory on another machine, and then 'extracted' it to / on the new machine.
But now everything (and I mean everything) is owned by "matt", my user on the intermediate machine. I rebooted the system to get it to boot the new stuff without realizing this, and now it's not booting... Since /etc/passwd, /bin/bash, etc. aren't owned by root.
Any tips on where to go from here? As I said, it's a brand-new machine (VM, actually: even better!), so re-imaging is a solution, but I'd still have to copy the data back over. Reinstalling is a pain, though, so if there's a simple fix, I'd prefer that. I've got pygrub installed as a boot manager for the VM, but I can't for the life of me figure out how to get into single-user mode. (Nor am I sure that a "cd /; chown -R root" will have the desired effect?)
