Trio3b

Upgrading pclos2009 (2.6.26 kernel) to 2.6.27 for better wireless support. New kernel fails to boot and leaves me w/blikning cursor. Upon using the good kernel to boot dmesg and /var/log messages seem to only record the successful boot and not the failed one. How can I record the failed boot to post here and correct what's wrong. Was looking for a boot log or something but no luck. Got the kernel from correct repos ( I think) and followed all instructions for upgrading. Any help appreciated. [Moved from Software - arctic]

Three different machines. Just found and copied a beatup and scratched vectorlinux disc successfully using dd. The deb discs are pristine but were not originally burned by me but both were kept in a warm location. I did try several block size options but no go on the debian discs. Here is messages right after trying the debian dd: Oct 14 09:40:01 laptop crond[13443]: (root) CMD ( /usr/share/msec/promisc_check.sh) Oct 14 09:41:01 laptop crond[13521]: (root) CMD ( /usr/share/msec/promisc_check.sh) Oct 14 09:42:02 laptop crond[13597]: (root) CMD ( /usr/share/msec/promisc_check.sh) Oct 14 09:43:01 laptop crond[13696]: (root) CMD ( /usr/share/msec/promisc_check.sh) Oct 14 09:43:03 laptop kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK Oct 14 09:43:03 laptop kernel: sr 1:0:0:0: [sr0] Sense Key : Medium Error [current] Oct 14 09:43:03 laptop kernel: sr 1:0:0:0: [sr0] Add. Sense: L-EC uncorrectable error Oct 14 09:43:03 laptop kernel: end_request: I/O error, dev sr0, sector 64 Oct 14 09:43:03 laptop kernel: Buffer I/O error on device sr0, logical block 16 Oct 14 09:43:03 laptop kernel: Buffer I/O error on device sr0, logical block 17 Oct 14 09:43:08 laptop kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK Oct 14 09:43:08 laptop kernel: sr 1:0:0:0: [sr0] Sense Key : Medium Error [current] Oct 14 09:43:08 laptop kernel: sr 1:0:0:0: [sr0] Add. Sense: L-EC uncorrectable error Oct 14 09:43:08 laptop kernel: end_request: I/O error, dev sr0, sector 64 Oct 14 09:43:08 laptop kernel: Buffer I/O error on device sr0, logical block 16 Oct 14 09:43:17 laptop kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK Oct 14 09:43:17 laptop kernel: sr 1:0:0:0: [sr0] Sense Key : Medium Error [current] Oct 14 09:43:17 laptop kernel: sr 1:0:0:0: [sr0] Add. Sense: L-EC uncorrectable error Oct 14 09:43:17 laptop kernel: end_request: I/O error, dev sr0, sector 64 Oct 14 09:43:17 laptop kernel: Buffer I/O error on device sr0, logical block 16 Oct 14 09:43:31 laptop kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK Oct 14 09:43:31 laptop kernel: sr 1:0:0:0: [sr0] Sense Key : Medium Error [current] Oct 14 09:43:31 laptop kernel: sr 1:0:0:0: [sr0] Add. Sense: L-EC uncorrectable error Oct 14 09:43:31 laptop kernel: end_request: I/O error, dev sr0, sector 64 Oct 14 09:43:31 laptop kernel: Buffer I/O error on device sr0, logical block 16 Oct 14 09:44:01 laptop crond[13774]: (root) CMD ( /usr/share/msec/promisc_check.sh) Oct 14 09:45:01 laptop crond[13850]: (root) CMD ( /usr/share/msec/promisc_check.sh) Anyway I have some releases on cd that are no longer available but really need to clean work area and would be nice to archive to HD, but I think this is going to be hit and miss. One thing I did notice is that the deb discs don't show up in the kde GUI notifier when inserted. I'm going to try more discs on several machines over the next few days and see if any patterns emerges. Thanks

UPDATE: bigger problem. Just got lucky with that 1 distro. That's why no reply post. Now 6 months laterI have tried 5-6 different distros on different brand media all in pristine condition. tried k3b and dd on two laptops running pclos2009.1 and one desktop with mdv2008.1. Not likely all three optical drives are bad. All discs are in pristine condition and I know they work b/c have done installs with them and md5 checks out. K3b copies about 30-100mb then stalls with "problem reading sectors" dd as root and as user also exit after partial reading with input/output error. I can see the iso file created in the destination folder but never complets. I have only been able to copy 2 cd isos out of about 15 discs. All drive lenses been cleaned. Does K3b have problems reading .isos that were burned on a different optical drive than the one trying to read it? Or with reading sector sizes of 2048? Read somewhere that cd-rw drives read isos better than cdrom drives. All three are dvdrom drives. Any ideas appreciated.

UPDATE: did two things 1. commented out the entry for sdb1 in fstab 2. moved mount point to /media/external05 and am now able to mount and unmount as user via GUI. Did this all NOT work via MCC because those tasks are done as root? We'll see how it goes after several reboots. THanks

have setup this drive now as sdb1 ( this is how MCC and fdisk identified the drive when plugged in) followed suggestions and via console made directory called external in /mnt, mounted at /mnt/external, then changed ownership of /mnt/external05, gave 777 permissions and was able to create some new directories and even store an .iso there. These directories and tasks were allowed by user However mounting and unmounting are only allowed by root in console. if user tries to mount or unmount via GUI in Konq system://media mount error: permission denied unmount error: device to unmount is not in /media/.hal-mtab so is not mounted by HAL Had to add drive to fstab manually and added the user to defaults but still no go. # Entry for /dev/sda1 : UUID=a26e688c-b021-4a85-9e90-660f401f51e5 / ext3 defaults 1 1 none /dev/pts devpts mode=0620 0 0 # Entry for /dev/sda6 : UUID=e698be64-8041-42b1-bf44-0dd11c5dc9f9 /home ext3 defaults 1 2 # Entry for /dev/sdb1 : UUID=a726a667-ea41-4d99-b46b-c9241a9ba709 /mnt/external05 ext3 defaults,user 0 0 none /proc proc defaults 0 0 none /tmp tmpfs defaults 0 0 # Entry for /dev/sda5 : UUID=8de447ca-28a3-41be-b36c-1b2d1e4985af swap swap defaults 0 0 Ideas? Thanks

It appears that I may have tried to experiment months ago with two partitions on hdb, one large for backup and one smaller for trying running Firefox cache into another disc but I guess I gave up on that and reformatted the disc into one partition and reinstalled with mount point under /storage. Trying to do too many things at once. Thanks for the help

Removed all references to /dev/sda in fstab to clean things up as I have disconnected the external drive for now. After futzing about on something else I come back to the desktop and here is fstab: # Entry for /dev/hda1 : UUID=6ec45ab5-4b17-4e5e-81d6-6cd4caf22b84 / ext3 relatime 1 1 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /backup ext3 defaults 1 2 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /cache ext3 defaults 1 2 # Entry for /dev/hda6 : UUID=5a211a14-e29e-4616-9c3f-39062081a788 /home ext3 relatime 1 2 /dev/cdrom /media/cdrom auto umask=0,users,iocharset=utf8,noauto,ro,exec 0 0 none /proc proc defaults 0 0 none /proc/bus/usb usbdevfs auto,devmode=0666 0 0 # Entry for /dev/hdb1 : UUID=6cada88c-a47c-492c-a036-64f0d22a226f /storage ext3 defaults 1 2 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /media/hd ext3 defaults 0 0 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /media/hd2 ext3 defaults 0 0 UUID=26980d91-2300-4653-a0ad-3482db8500f3 swap swap defaults 0 0 I wondered why I was still able to access hdb when both partitions were commented out in fstab posted earlier. Now I see that a "new" entry has appeared in fstab compared to the original that I cut and paste directly from konsole in this thread and in my thread in hardware forum. Where did that line come from? That line was not there earlier! what the h*ll! No wonder hdb is working! confused Anyway, I would still like to clean up all those hdb partitions in fstab. we have starrted using hdb as backup so I would prefer not to have to start from scratch in adding hdb to fstab. Any ideas on which hdb entries can be deleted? I guess I could save this fstab as fstab.old and then write a new fstab removing all the hdb entries EXCEPT the one in orange and see if all is OK? Thoughts? Thanks

I have removed all references to sda (which is what the external drive shows up as) from fstab and this has cleaned things up a bit. The multiple hdb entries is also distracting me but I have started a different thread in software for that. I am going to skip using MCC and try your method, but doesn't mdv mount these devices under /media? I notice that is where it shows up in the storage media. Thank you

mdv 2008.1 installed on 80 gb hda: It appears I've made a mess of fstab. Everything is working but I am trying to add an external HD (IDE 3.5 in a USB enclosure) and looking at fstab gives me cause for concern as explained in a sepaprate topic in hardware forums. Approx 4 mos. after building this pc, decided to add another internal 80 gb HD on IDE1 slave position as a backup for hda. Not a clone, but just a place for users to have their own backup folder on a separate drive. I must have tried installing via MCC and mounted it at /storage when MCC asked for mount point. I then created two directories, user1 and user2 and set permissions so each user can only access his/her folder, then on each desktop created a link to each users backup folder on the /storage location (URL). This seems to have worked out for the past 6 mos. Today upon viewing fstab to add an external drive I was surprised at this: # Entry for /dev/hda1 : UUID=6ec45ab5-4b17-4e5e-81d6-6cd4caf22b84 / ext3 relatime 1 1 # Entry for /dev/sda5 : UUID=bde2240a-c04f-4eba-a83a-05f553522a71 /ext_user01 ext3 relatime 0 0 # Entry for /dev/sda1 : UUID=7963c1c7-7e85-490b-8eac-afa4742ab4aa /ext_user02 ext3 relatime 0 0 # Entry for /dev/sda1 : #LABEL=WD-320external /external ext3 defaults 0 0 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /backup ext3 defaults 1 2 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /cache ext3 defaults 1 2 # Entry for /dev/hda6 : UUID=5a211a14-e29e-4616-9c3f-39062081a788 /home ext3 relatime 1 2 /dev/cdrom /media/cdrom auto umask=0,users,iocharset=utf8,noauto,ro,exec 0 0 none /proc proc defaults 0 0 none /proc/bus/usb usbdevfs auto,devmode=0666 0 0 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /media/hd ext3 defaults 0 0 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /media/hd2 ext3 defaults 0 0 UUID=26980d91-2300-4653-a0ad-3482db8500f3 swap swap defaults 0 0 Somehow I must have bolloxed up the hdb install and commented it out then tried again with another method, but if you look, it appears that all hdb entries are commented out. It appears that the device partitions are the same UUID but are mounted in different locations. I did this using MCC. Maybe I should mkdir usbdisk in /mnt or /media 1. How is it that partitions are available if they are commented out? 2. How can I start to clean up fstab to be rid of unused partitions and how to figure which ones are necessary? 3. Yes there are sda1 (twice) and sda5 but that issue is in another thread. Any help appreciated. Thanks

Upon reflection, I believe hdb was added AFTER the initial OS installation on hda and my problem is coming from the inability to add additional devices to fstab properly. Had hdb (internal) and the new external device been present while installing mdv2008.1 on hda, I think all would be fine. I just assumed that the "new" external drive (IDE to USB) would be seen as a USB storage device just like any flashdrive. Also just spotted aussiejohn's post about automounting an external just like a flashdrive. That is basically what I would like to do. Remembering to unmount and safely remove the device partition(s) is doable but in an office setting with someone unfamiliar, it could create problems.

MDV 2008.1 on hda with internal hdb mounted at /storage for backup. Now would like to add exteranl HD. purchased WD320gb internal 3.5" with intention of installing into another PC but decided to go external. Bought enclosure (IDE 3.5" to USB) installed and plugged it in. USB flashdrive icon appears, and opens in konq, but cannot create any directories. "access media/disk denied" Strange, but in MCC disk management volume shows up as ext3 but I don't remember formatting it.(could be wrong, I bought it several months ago, tinkered and then left it alone). Not likeley it came from store formatted in ext3. Anyway, then went to MCC and it shows up as sda1 ( I regularly plug in little 2gb flashdrives and they also show up as sda1), labeled it and told MCC to save to fstab and tried again to access. No go. So I just commented it out in fstab and tried to just hotplug again as normal user, but access still denied. Pretty sure I'm missing something simple (permissions) but just not seeing the forest for the trees. internal hdb is also owned by root but am able to create a separate backup folder for each user with no problem. Why not with this external IDE to USB HD? Update: plugged external into lappy, went to MCC and formatted as ext3. (for some reason the dessktop earlier saw it as ext3 but had not been formatted so formatted entire part as ext3) . Then back to the desktop. Now went to MCC again and created two ext3 partitions for both users mounted as /ext_user01 and /ext_user02 each 150gb in size. Then created a directory in each part, one for user01 and one for user02 via CL and assigned ownerships for each user, but still can't mount the device as user. here is fstab: # Entry for /dev/hda1 : UUID=6ec45ab5-4b17-4e5e-81d6-6cd4caf22b84 / ext3 relatime 1 1 # Entry for /dev/sda5 : UUID=bde2240a-c04f-4eba-a83a-05f553522a71 /ext_user01 ext3 relatime 0 0 # Entry for /dev/sda1 : UUID=7963c1c7-7e85-490b-8eac-afa4742ab4aa /ext_user02 ext3 relatime 0 0 # Entry for /dev/sda1 : #LABEL=WD-320external /external ext3 defaults 0 0 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /backup ext3 defaults 1 2 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /cache ext3 defaults 1 2 # Entry for /dev/hda6 : UUID=5a211a14-e29e-4616-9c3f-39062081a788 /home ext3 relatime 1 2 /dev/cdrom /media/cdrom auto umask=0,users,iocharset=utf8,noauto,ro,exec 0 0 none /proc proc defaults 0 0 none /proc/bus/usb usbdevfs auto,devmode=0666 0 0 # Entry for /dev/hdb1 : #UUID=200e5a69-099c-4fb5-98d6-9c72a2d91f0c /media/hd ext3 defaults 0 0 # Entry for /dev/hdb5 : #UUID=bd4b11b7-b34d-47af-85a2-9c3b2be84d70 /media/hd2 ext3 defaults 0 0 UUID=26980d91-2300-4653-a0ad-3482db8500f3 swap swap defaults 0 0 The two partitions I'm working with are sda1 and sda5 on the external. If it helps my root part is on hda1, home on hda6, sda1 (#LABEL=WD-320external /external ext3 defaults 0 0) is where I first installed the external but then commented it out. please disregard the commented out hdb partitions as that is another mystery as to why it's still working. Will save that for another post. Apparently I decided to mount sda1 and sda5 on /ext_user01 and /ext_user02, but the these are mounted on /dev. Is there a better way? I am going to leave this external plugged in as backup most of the time BUT NOT RUNNING most of the time, just weekly backups and I just want it to mount automatically when a user turns it on to backup. then when they shut it off I guess it will unmount just like a flashdrive. I also purchased a larger flashdrive and did not have these issues. Why can I access the f/d as user but not the external HD? Any help appreciated Thanks

MDV 2008.1 Konqueror folder timestamps seem random. Folder creation date seems to be modified to the date when moving folders to other devices (e.g. flashdrives, other PCs) Access and modification dates seem not to be correct either. For example I have a folder that I am positive was created 5 years ago but creation date is today 03/29/10. stat command also gives this date. Any fix or update for this? Any help appreciated.

I'm going to try this on a non-critical install for practice but can't get to it for a week or so. Thanks

I think BIOS psswd and removing CDROM from boot selection is the way to go for now to at least plug up those holes. Not experienced enough to mess with GRUB yet but also sounds like a great option, because I think you can still pass "single" to the kernel at the bootsplash or the option to enter the console at the login screen so yes protecting GRUB is something to look into. Thanks

Hmmm . . these sound like viable options. Will give them a go. Thanks

Yes the LiveCD would be a way in but it does require somewhat of a learning curve for the newb and I'm not sure this person would pursue this. I try to make sure that no really sensitive data such as SS# or credit card info is on the PC at all anyway. Just being paranoid I guess. Encryption is definitely an option but as mentioned I'm new to that and want to do it right so I don't know if I have time for that right now. Years ago I did install a minitoggle switch hidden on the back panel of a tower to disconnect power to the CDrom drive to keep the kids from using the drive. I guess I could do that again. The BIOS password option is a good one as well and I have used that in the past, but yes, a business environment requires a certain amount of fluid access to the PC and its files. I think maybe I'll go back to mowing lawns and keeping client info on a pocket notebook like I did as a teen. :P Thanks for the replies.

Running mdv2008.1 I have hired an assistant to help in the office and have already locked permissions on files/folders I don't want them to have access to but I remembered there is a way to reset root password by entering single user mode at boot and typing passwd. It will then prompt to enter new password and confirm without asking for the old / original password. The likelyhood that this person would go thru the trouble to learn how to do this is slim as they are unfamiliar with Linux but I want to cover all angles. 1. Is this correct? 2. How can I prevent other/unauthorized users from doing this? I read up on /etc/passwd and etc/shadow but not sure I'm understanding. Looked into etc/passwd and it lists user IDs and etc/shadow contains the password in encoded or encrypted form but you need root password to view /etc/shadow which is a good thing, but this doesn't seem to address the ability of an unauthorized person from entering single user mode at boot and changing the password using the command passwd. I don't think I need to protect against cracking the root password per se (or maybe I do), but more against someone changing it. Any other security tips appreciated. I don't think I'm ready to encrypt folders and files yet. One thing at a time for me right now. Thanks

Will give that a go. Thanks gentlemen

Most of us download an iso of a distro onto HD and then burn to a bootable disc which we use to install the OS onto another / different HD. I want to go in reverse. Take a distro already burned to CD and either put it back into an iso file or archive it somehow onto a spare HD. Why? 1. Running out of room in my shop and want to toss out / donate many of the older distros on CD that I don't use anymore but MIGHT want to tinker with someday. 2. Because some of the older release versions that I have on CD are either not available / hidden /difficult to find and some of these older unavailable release versions are actually better than some new releases. 3. Also, because I did much of my d/l onto a little 20g HD and had to delete the iso for space sake, but now I have access to much larger HD Is this doable ? Can the process of burning an iso file to disc be reversed? Does this work for a LiveCD? Curious. Thanks

That's as good an explanation as anything. Will comment out again and see what happens. thnx

As mentioned, hdb used for backup went bad and I had to comment out hdb partitions in fstab in order to boot mdv2008.1 (on hda) . Now having trouble mounting flashdrives so poked into fstab and guess what? hdb1 and hdb5 partitions are listed in fstab again! What's going on? hdb is out and in a different PC!! Any ideas? (not about the flashdrive, but about the hdb entry. UUID's are the same as the original hdb partitions). Thanks

running mdv 2008.1 40gb backup drive (hdb) has gone bad. Tried the freezer trick, no luck. hdb (with partitions hdb1 and hdb3) is now out of the PC and in another machine running pclos 2007 and trying to recover using ddrescue. Have set up a larger partition (hda7) specifically to contain files on the unreadeable hdb1 partition. Once ddrescue finishes, hda7 won't mount and reads similar errors as hdb1 : either bad permissions or bad superblock, fs, etc. Anyone experienced with ddrescue? Thanks

Well, /home is on a different partition and I am mostly backed up onto a second HD via grsync/rsync as well as on discs. As an aside, I am leary of using drakbackup as it has filled /var to the point of non-booting if you're not careful (anyone know if this has been fixed with a warning?), but I've never restored from drakbackup. How does it work? 1.Can the compressed drakbackup file be placed on another partition, harddrive? 2. If there is a catastrophic failure and you reinstall MDV fresh, including a new /home partition, can you invoke the drakbackup from a file that was created by the now non-existent user? I notice drakbackup names the files "backup_base_user_username". Would the newly installed user (on a reinstall or newly created user) have to have the same name as the original backup file? Having said all this, I have never had updates bork my system, but that was with 10.2. Just being cautious. Any help appreciated.

Preparing to perform updates on mdv2008.1 and want to make sure that a full update won't cause problems as this PC is loaded with my data. This is an update of 2008.1 not an upgrade of 2008.1 to something newer. Any issues I should be aware of? Also, in the past, I just select updates for packages that are system/security related and usually do not update software that i don't use. Is individual safer/better than update all? Thanks for any help

I have a spare VIA C3 with a 320gb HD sitting around that I would like to setup as a poor mans NAS/backup server. At some point I will look into RAID but just can't do it now. Would also backup to DVD as secondary. Files are mostly text, spreadsheet and PDF files. MDV 2008.1 and PCLOS 2007/2009 workstations are on the LAN. I am less concerned with the actual backup software and more with the setup of my LAN and the headless configuration of the server itself. 1. Will install the OS on a 20gb hd and use the 320 gb hd as data storage 2.From what I've read, this older VIA C3 (800mhz) with 256 or 512 mb RAM should be fine for this usage. 3.Curious about CentOS for the OS. Any thoughts or should I just stick with what I know MDV/PCLOS? 4. I want to make it headless. I believe that I can install the OS, set it up the way I want with ssh and then disconnect the monitor and KB (checking BIOS to see if there will be any no-KB faults). At this point I just want to turn it on once a week and use it to backup business files or MDV/PCLOS system files. 5. I want to keep the GUI (runlevel 5) intact so all users can just ssh in to it from the other workstations, is this correct? 6. Will use grsync, Unison to backup and sync files. I can learn to cron for auto backups. Have looked into openfiler /others and they seem way overkill for my needs. 7. At this point there will be no Windows on this LAN, maybe later. This is my current setup: internet <-> MDV 2008.1 as gateway <-> hub/switch <-> all other computers including PCLOS, occasional lappy, and now add the VIA server. Any thoughts appreciated