Poking around reveals that one Dave Chinner, Principal Engineer -- SGI Australian Software Group, submitted a series of patches to deal with preventing the flushing of stale inodes. I'm trying to figure out just how relevant those patches are to the stability of XFS, especially to pre-Lucid Ubuntu systems: are they in danger of data loss? Some quick search results thrown together in an effort to asses this issue:[URL]..Quote:
Dave Chinner (9):
xfs: Don't flush stale inodes
xfs: Ensure we force all busy extents in range to disk
xfs: reclaim inodes under a write lock
xfs: Avoid inodes in reclaim when flushing from inode cache
I have recently upgraded my Debian unstable to the new custom 2.6.34 kernel. The system is unresponsive and top shows two processes (flush-8:0 & flush-8:16) taking up most of the cpu time.
I decided to take an old Gateway that I bought off a guy cheaply and turn it into a file and web server.I purchased copies of Debian 5.0.4 i386 disks (31 in all) on the advice of a friend, the disks weren't expensive, but now that Ive installed all the disks, I'm having a variety of errors
[443.110940 end request: I/O error, dev hde, sector 76021855 [443.111074] EXT3-fs error (device hde1): ext3_get_inode_loc: unable to read inode block - inode=2375715, block=9502724 INIT: cannot execute "/sbin/getty"
policy is to backup mysql with mysql-zrm.However at a certain stage it hangs forever. This is at the "flush logs". I tried this manually and it gave the same result. Even after restarting mysql and the host. After some googling and trying I found out "flush tables with read lock" gives the same result. The tables seem to be MyIsam. I tried with a mysqldump on one server and restore it on a test vm. I used the same config and flush logs still hangs. ALso I tried to change some configuration directives... but with the same result
Edit: btw, I checked the logfile and didn't found anything (/var/log/mysqld.log)
Edit2: I also did myisamchk -s *.MYI (in all direcoties with db files;actually did it with find command) and it did not return anything so datafiles seem ok.
I find that the FC 15 Xfce spin will save the user's current session except that it doesn't save what is being done with the Lyx editor. Under FC 15 Gnome 3, I found that after I added the minimize-maximize-close buttons to the top of windows, that some of the windows in Lyx (such as the one that appears for View as DVIPS) didn't respond to those buttons. (I don't remember whether Gnome 3 saved the Lyx session and I've gotten rid of Gnome 3 now.)
I was using gparted from a live usb to resize an ext4 partition and it failed while running resize2fs. The error it gave was
Code: resize2fs: The inode is from a bad block in the inode table while trying to resize /dev/sda5 please run 'e2fsck -fy /dev/sda5' to fix the filesystem after the aborted resize operation.
I get the error "could not open obex://some_long_MAC_address...please do something..." when I try to browse my samsung using bluetooth.
My phone's service list indicates all but one service item(modem) are present. I can send stuff to it but not the other way round. Someone told me that debian distros have bluetooth problems.
Also, when compiling kernel 2.6.36.5 and 2.6.37-etc, I get the undisputed error concerning the 'lguest' feature. How do I turn it off?
When I upgraded to 10.4LT I agreed to something that stopped snort, after days decided to just re-do with new snort version. Used bodhi.zazen's MySql instruction version (which is what I used in the past) Everything went pretty well except for figuring out that I needed to delete all the lib_sfdynamic_preprocessor_example?? files (I also deleted all the lib_sfdynamic_example?? files too just to be safe). Used my original Oinkmaster with updated rules version and downloaded the emerging threats too (as I had in the past) and snort won't run with some of the emerging threat rules because it's lookning for snortsam (fwsam). I read up and snortsam looks like a good idea (if I'm wrong somebody just let me know)
if this seems dumb, but I really don't understand, the snortsam directions are HORRIBLE, the snortsam src looks like a windows file when unpacked with all the .dll files(but they say for all OS's), it builds but you need to copy the binary to /usr/local/bin (what in ubuntu would be a binary?).
the snortsam-patch-2.8.tar.gz won't unpack and the Snort 2.8.6 patch is a file, not a package (have no clue where to put it or what to call it if I got the 2.8.tar.gz to unpack so I could build it)
Either Verizon has been hacked or they have installed something that should be a security concern.I have found that following Javascript "injected" into most webpages. I know it is added via a Verizon (wireless broadband) connection because a) it does not happen when I use Embarq DSL, and b) is happens to webpages that I have created (and I certainly did not add the Javascript!). So, here it is:
Well, that's my story. My concern is that I was able to view some private information (via a backend login) using the modified link, example: http:// 62.0.5.133/www.somedomain.comm/login/myaccount.php!! (this displayed private info!)I have since installed Privoxy to remove the offending Javascript before it gets actuated by my browser.
I work in a relatively small organisation of about 30 people (but with a complex network) and we've been looking to move our firewall to Microsoft's Threat Protection Manager on a mostly Windows network. I've been thinking we should have an IDS/IPS inside the firewall and I've been thinking about Snort in NIDS mode but have some basic questions:
1. Can anyone recommend a good web GUI for Snort?
2. Is it advisable to run both on the same machine? (Both from a POV of security and resources.)
Two days ago we started to receive the following message:
/etc/cron.daily/chkrootkit: The following suspicious files and directories were found: /lib/init/rw/.mdadm /lib/init/rw/.ramfs /lib/init/rw/.mdadm INFECTED (PORTS: 4369) You have 2 process hidden for ps command chkproc: Warning: Possible LKM Trojan installed
And about at the same time (a day before that) we have set up new rules for the queueing disciplines using 'tc' on our Debian lenny box (these rules are for some of the experiments we are carrying out). I have ran the chkrootkit manually and this message (as above) keeps appearing, while the rkhunter tool does not complain about these items. Could there be a connection between setting up the new qdisc's and the chkrootkit "INFECTED" messages?
Slackware64 13.0 I have a website that has been migrated to a new server. After a few days I'm still pointing at the old server and the hosting company have recommended that I flush my dns cache. A quick google for this in linux seemed to recommend restarting the nscd daemon but nscd -g tells me it's not running anyway. I connect to Virginmedia via wired ethernet to a Belkin N1 router so is there a way I can do this I wonder, either via Virginmedia or Slackware?
I have an HP printer for my Lenny which has worked for some year. But I don't remember what method I used to install it. So this is one piece of the puzzle that I can't see. But like I said the printer works. One day I accidentally printed more than I had papers in the printer-machine. Then I kind of stacked a lot of print jobs in the queue out of frustration. So whenever I reboot the PC/Lenny then it waste some paper by printing things that got stuck in the printer queue. It's not very environmental this weird behavior.
So next time this happens how do I flush the Printer queue so Lenny doesn't remember what happened before the reboot? I followed these instructions earlier but it only switched one weird behavior with another weird behavior. So it didn't work for my Lenny, and I couldn't find any better solutions on the Internet. [URL]...
I believe I have unwanted ' characters left in a 9 element character array that are causing subsequent operations with it to fail. I see wildly differing views on the web on the proper way to flush 'em. It's clearly not as simple as it would appear at first sight. What's currently the best (or else "least deprecated") method?
why does my virtual machine freezes when I flush iptable rules. i tried to install virtual machines 3 times and every time I flush iptables on host, virt machine freezes down. What can be the issue? Is it with the host installation or something else?
Is it possible to remove the "flush" flag when mounting removable disks in KDE4 without recompiling KDE4? Can it be done in some config file(s)? Thanks!
I am attempting to write a backup script that will do the following:
1) lock and flush tables on a mysql db 2) dump the db to a file 3) unlock the tables 4) rsync the file to offsite storage
It all seems to be going well. However, obviously I don't want to setup ssh to the storage server on another network as the root user without a password. so I am attempting to su as the backup user inside of the script but when I try to run the script everything happens as it should until I try to so.. then it jumps out of the script .. akss me to login as the backup user.. proceeds to rsync to the offsite storage it does all this and then resumes execiting the script. it is not going to be setup as a cron job. it will be executed manually. assuming that is the case, how can I get the script to run without prompting for a password?
Here is what I've come up with so far... assuming that the script is run as root and the identity of the backup user will need to be assumed inside the script without perstering the user to enter the backup user's password.
I am running the x64 version of Ubuntu 9.10. For the past 2 days(haven't installed any new programs within the same time frame). I've started to encounter problems after I lock my computer, (ctrl+alt+l), then come back and unlock it. After unlocking, I am unable to open any programs.
Attempts to open programs include: panel icon buttons, application menu buttons(including opening nautilus through the places menu), alt+F2, and using gnome-do. I also can't use any of the buttons in the user menu to the far right to log off or shut down the computer and am forced to hold the power button to restart my laptop. I am able to use programs that are already open with no problems.
Also, If i lock the laptop again, the screen darkens as expected, except, when i move the mouse to bring up the login dialog all i see is short flickers of my mouse as it moves around a black screen.
My most recent synaptic log includes:
Quote:
Commit Log for Sat Feb 27 17:09:49 2010 Upgraded the following packages: foomatic-filters (4.0.3-0ubuntu2.1) to 4.0.3-0ubuntu2.2 google-chrome-unstable (5.0.322.2-r38810) to 5.0.335.0-r39561 lernid (0.5) to 0.6 sudo (1.7.0-1ubuntu2) to 1.7.0-1ubuntu2.1
Recently while at school, my trackpad stopped working completely. When I log out to my login screen it works, and as well as kde. I use compiz btw in gnome.[SOLVED I THINK] --> My boot splash doesn't show up until its fully loaded and won't show in the correct resolution.Every so often, the UI flips out and won't let me click menus or text fields or use my keyboard at all.
The trackpad is a setting i bet. BUT I looked everywhere for a setting that is making it not work.[SOLVED I THINK] --> The splash is ok, I'm not worried about it, but I wished it looked nicer on my computer D:The UI freak out is annoying and trouble some as it happens randomly in use.
I am on a laptop:
HP Pavilion dv9000 Intel Centrino Core 2 Duo 1.73GHz NVIDIA GeForce Go 7600 2GB of memory
ps - I am not a fan of HP, Intel, or NVIDIA, this laptop was free so not complaining about not having AMD or ATi or another laptop brand
I've been using ubuntu almost exclusively for more than a year. I am now running 10.04. I was trying via launchpad to post some bugs I've run into in Ubuntu. None of the bugs relate directly to a certain process. And if they do, I have no idea how to find out their PIDs.
1. Bug one. For some strange reason, my mouse is red only when hovering over a firefox window. All the other time, its white. I have tried to use the built in cursor changer thing, and it seems only capable of changing the cursor for the desktop. Not this strange, magical red firefox cursor. It starts to mess with ya.2. Not really a bug I guess, just a mild annoyance. In windows 7, you have a good deal of play when it comes to where you want to position your icons on the desktop. Not so with Ubuntu 10.04. Its more like windows 95 when it comes to that. Is there some way to solve this? You have like an inch between places you can move icons to. Makes the desktop look ugly, and I would just feel more at home if I could position them correctly. The pic is an accurate representation of this. On the far left, that is as far as the icons can go. And if you want them near each other, they have to sit way too close, or way too far apart. (oh and that pic was before I attempted to change the cursor lol. Now its white when not hovering over FF, red when over FF. Now I'm stuck in that configuration lol)
How does a filename is mapped to its inode ??? If I want to make our own system call and use a filename as argument how can I get its inode ,if I want to use some of member of inode structure in code. Basically I want to get the fd of the file.
In my project, there is a requirement for adding one more attribute for file properties which keeps track of previous file copies(Copy history). So how can we accomplish this? can we add/store this information in inode structure?
I have a problem with installing ubuntu 11.04 I get this message: Code: Hspell: can't open /usr/share/hspell/Hebrew.wgz.sizes It's get stuck with the updating time from internet servers process but I can't find a way to report it to launchpad.
I did a clean install of Lucid, and almost everything video related is a little sluggish. First of all, Compiz is really slow compared to how it was in Karmic. I don't even use window animations, just the Scale and a couple other effects, like desktop cube. Scale is really slow, and sometimes it doesn't work. I have to start compiz again to get it working. Desktop cube actually froze a couple of times half way through a rotation. Rotating the cube is much slower as well.
Also, I've been having a lot of trouble with HD video playback. It's not just the 100% CPU load that's bugging me, but every 3-4 minutes the playback skips frames, and the image becomes all glitchy.
Flash playback also uses a lot of CPU, I can't even watch ..... videos in fullscreen most of the time. And I'm always opting for the lowest quality possible btw.
I have a dual core AMD CPU and 8600GT nvidia graphics card, so it's not a hardware issue.
