Re: Run out of inodes
--- Biran, Yahav (Yahav) [EMAIL PROTECTED] wrote: df -I show: [EMAIL PROTECTED] mpower]# df -i FilesystemInodes IUsed IFree IUse% Mounted on /dev/cciss/c0d0p6 262144 10442 2517024% / /dev/cciss/c0d0p1 26104 54 260501% /boot none 1013170 1 10131691% /dev/shm /dev/cciss/c0d0p23989888 616728 3373160 16% /export/home /dev/cciss/c0d0p7 131616 20 1315961% /tmp /dev/cciss/c0d0p33842720 127472 37152484% /usr /dev/cciss/c0d0p8 131616 131616 0 100% /var Therefore I can't run any rpm nor up2date. How one can clean up some inodes? You can remove some files under /var. Or you can backup /var somewhere, unmount it, create on /dev/cciss/c0d0p8 new filesystem with more inodes, and then restore /var Valery. Valery Internal Virus Database is out-of-date. Checked by AVG Free Edition. Version: 7.5.503 / Virus Database: 269.17.1/1183 - Release Date: 12/13/2007 9:15 AM To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Run out of inodes
Few years ago a Linux system, which I used, ran out of inodes once in a while. The reason was that a directory in /var (I do not remember its name) got filled by thousands of zero-length files, due to a botched error recovery attempt by some daemon. The cure was to delete all those files. I would suggest that you look for obsolete archive files in /var/log and for other obsolete stuff in other /var subdirectories. Once those files re deleted, you should be able to free several inodes. --- Omer On Wed, 2007-12-26 at 09:12 +0100, Biran, Yahav (Yahav) wrote: df -I show: [EMAIL PROTECTED] mpower]# df -i FilesystemInodes IUsed IFree IUse% Mounted on /dev/cciss/c0d0p6 262144 10442 2517024% / /dev/cciss/c0d0p1 26104 54 260501% /boot none 1013170 1 10131691% /dev/shm /dev/cciss/c0d0p23989888 616728 3373160 16% /export/home /dev/cciss/c0d0p7 131616 20 1315961% /tmp /dev/cciss/c0d0p33842720 127472 37152484% /usr /dev/cciss/c0d0p8 131616 131616 0 100% /var Therefore I can't run any rpm nor up2date. How one can clean up some inodes? -- MS-Windows is the Pal-Kal of the PC world. My own blog is at http://www.zak.co.il/tddpirate/ My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with which I may be affiliated in any way. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Find process id of background ssh?
Hi List. I'm writing a script to automate some system maintenance tasks, and I want to connect over SSH to several remote computers and do stuff on them. I'm using ssh -f to background ssh so I can run the same operation on multiple machines in parallel, otherwise it will be too slow - the maintenance job may take up to a few minutes to run and the script is not supposed to be fully automatic: a human is to monitor the process. But I don't want just to fire and forget the SSH processes - I want to exit from the script only when all the SSH processes have completed. I can do that by monitoring the process ids of the background SSH processes, if I could know them - which I'm having a difficult time detecting. I'm writing in bash, and optimally it would be something like this: for server in 1 2 ...; do ssh -f [EMAIL PROTECTED] 'run maintenance task' pids=$pids $(getSSHpid) done while kill -0 $pids 2/dev/null; do echo Waiting..; sleep 1; done but I didn't manage to find a way to get the process id of the ssh process after it goes to background, other the 'ps'ing for it. How can I go about doing this? -- Oded Sorry for OT, but some kind of distributed shell seems me more suitable for this task - http://www.linux-mag.com/microsites.php?site=business-class-hpcsid=buildp=4658 = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Run out of inodes
df -I show: [EMAIL PROTECTED] mpower]# df -i FilesystemInodes IUsed IFree IUse% Mounted on /dev/cciss/c0d0p6 262144 10442 2517024% / /dev/cciss/c0d0p1 26104 54 260501% /boot none 1013170 1 10131691% /dev/shm /dev/cciss/c0d0p23989888 616728 3373160 16% /export/home /dev/cciss/c0d0p7 131616 20 1315961% /tmp /dev/cciss/c0d0p33842720 127472 37152484% /usr /dev/cciss/c0d0p8 131616 131616 0 100% /var Therefore I can't run any rpm nor up2date. How one can clean up some inodes? Internal Virus Database is out-of-date. Checked by AVG Free Edition. Version: 7.5.503 / Virus Database: 269.17.1/1183 - Release Date: 12/13/2007 9:15 AM To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Asterisk won't detect the line was hung up
On Wed, Dec 26, 2007 at 10:30:04PM +0200, Shachar Shemesh wrote: Both /etc/zaptel.conf and /etc/asterisk/zapata.conf attached. The provider is Bezeq. Tzafrir Cohen wrote: busydetect=yes Yes. I uncommented busydetect, busycount and busypattern (500,500), and the problem seems better now (though not perfect). I'll have to tweak those around a bit, I think. You don't really need to set busypattern. (and is the busypattern here 500,500?) Is polarity reversal used anywhere in Israel? (Actually: it is simple to test: enable debug loggin inAsterisk - in logger.conf, 'core set debug 5', and look at the messages you get when a call comes in. See if you see a message about polarity). I'm not sure I did what I was meant to do, but assuming I did, no, there was no mention of polarity. Right. Not supported here. We have to resort to busydetect. 1. Asterisk does not receive the caller ID from Bezeq. This is usually because the line doesn't have caller ID enabled. Check that with a caller-ID-enabled phone. Checked, and the line does have caller ID. That's not it. Quoting your zapata.conf here: callerid=asrecieved channel = 7 callerid=asrecieved channel = 8 You have a typo there: $ grep asrecieved /usr/lib/asterisk/modules/chan_zap.so $ strings /usr/lib/asterisk/modules/chan_zap.so | grep asre asreceived That is a typo I have made too often. 2. I cannot define a call group for outgoing calls (Zap/g1). Incorrectly-configured zapata.conf? Could you please post it? Attached are both files. Try: egrep '^(group|signalling|channel)' /etc/asterisk/zapata.conf You have your FXS channel (1) in group 0 and your FXO channels (7,8) in group 2. Try Zap/g2 instead of Zap/g1 . -- Tzafrir Cohen | [EMAIL PROTECTED] | VIM is http://tzafrir.org.il || a Mutt's [EMAIL PROTECTED] || best ICQ# 16849754 || friend = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Asterisk won't detect the line was hung up
Both /etc/zaptel.conf and /etc/asterisk/zapata.conf attached. The provider is Bezeq. Tzafrir Cohen wrote: busydetect=yes Yes. I uncommented busydetect, busycount and busypattern (500,500), and the problem seems better now (though not perfect). I'll have to tweak those around a bit, I think. Is polarity reversal used anywhere in Israel? (Actually: it is simple to test: enable debug loggin inAsterisk - in logger.conf, 'core set debug 5', and look at the messages you get when a call comes in. See if you see a message about polarity). I'm not sure I did what I was meant to do, but assuming I did, no, there was no mention of polarity. 1. Asterisk does not receive the caller ID from Bezeq. This is usually because the line doesn't have caller ID enabled. Check that with a caller-ID-enabled phone. Checked, and the line does have caller ID. That's not it. 2. I cannot define a call group for outgoing calls (Zap/g1). Incorrectly-configured zapata.conf? Could you please post it? Attached are both files. Thanks for your help. Shachar ; ; Zapata telephony interface ; ; Configuration file ; ; You need to restart Asterisk to re-configure the Zap channel ; CLI reload chan_zap.so ; will reload the configuration file, ; but not all configuration options are ; re-configured during a reload. [trunkgroups] ; ; Trunk groups are used for NFAS or GR-303 connections. ; ; Group: Defines a trunk group. ;trunkgroup = trunkgroup,dchannel[,backup1...] ; ;trunkgroup is the numerical trunk group to create ;dchannelis the zap channel which will have the ;d-channel for the trunk. ;backup1 is an optional list of backup d-channels. ; ;trunkgroup = 1,24,48 ;trunkgroup = 1,24 ; ; Spanmap: Associates a span with a trunk group ;spanmap = zapspan,trunkgroup[,logicalspan] ; ;zapspan is the zap span number to associate ;trunkgroup is the trunkgroup (specified above) for the mapping ;logicalspan is the logical span number within the trunk group to use. ;if unspecified, no logical span number is used. ; ;spanmap = 1,1,1 ;spanmap = 2,1,2 ;spanmap = 3,1,3 ;spanmap = 4,1,4 [channels] ; ; Default language ; ;language=en ; ; Default context ; ;context=default context=incoming ; ; Switchtype: Only used for PRI. ; ; national: National ISDN 2 (default) ; dms100: Nortel DMS100 ; 4ess: ATT 4ESS ; 5ess: Lucent 5ESS ; euroisdn: EuroISDN ; ni1:Old National ISDN 1 ; qsig: Q.SIG ; switchtype=national ; ; Some switches (ATT especially) require network specific facility IE ; supported values are currently 'none', 'sdn', 'megacom', 'tollfreemegacom', 'accunet' ; ;nsf=none ; ; PRI Dialplan: Only RARELY used for PRI. ; ; unknown:Unknown ; private:Private ISDN ; local: Local ISDN ; national: National ISDN ; international: International ISDN ; dynamic:Dynamically selects the appropriate dialplan ; ;pridialplan=national ; ; PRI Local Dialplan: Only RARELY used for PRI (sets the calling number's numbering plan) ; ; unknown:Unknown ; private:Private ISDN ; local: Local ISDN ; national: National ISDN ; international: International ISDN ; dynamic:Dynamically selects the appropriate dialplan ; ;prilocaldialplan=national ; ; PRI callerid prefixes based on the given TON/NPI (dialplan) ; This is especially needed for euroisdn E1-PRIs ; ; sample 1 for Germany ;internationalprefix = 00 ;nationalprefix = 0 ;localprefix = 0711 ;privateprefix = 07115678 ;unknownprefix = ; ; sample 2 for Germany ;internationalprefix = + ;nationalprefix = +49 ;localprefix = +49711 ;privateprefix = +497115678 ;unknownprefix = ; ; PRI resetinterval: sets the time in seconds between restart of unused ; channels, defaults to 3600; minimum 60 seconds. Some PBXs don't like ; channel restarts. so set the interval to a very long interval e.g. 1 ; or 'never' to disable *entirely*. ; ;resetinterval = 3600 ; ; Overlap dialing mode (sending overlap digits) ; ;overlapdial=yes ; ; PRI Out of band indications. ; Enable this to report Busy and Congestion on a PRI using out-of-band ; notification. Inband indication, as used by Asterisk doesn't seem to work ; with all telcos. ; ; outofband: Signal Busy/Congestion out of band with RELEASE/DISCONNECT ; inband: Signal Busy/Congestion using in-band tones ; ; priindication = outofband ; ; If you need to override the existing channels selection routine and force all ; PRI channels to be marked as exclusively selected, set this to yes. ; priexclusive = yes ; ; ISDN Timers ; All of the ISDN timers and counters that are used are configurable. Specify ; the timer name, and its value (in ms for timers). ; K:Layer 2 max number of outstanding unacknowledged I frames (default 7) ; N200: Layer
Re: [YBA] NIS vs LDAP
Hi All, All of you consider Active directory as an LDAP server, this is not truth, AD is a kerberos environment that LDAP served as a backend (the first AD server is authentication server, TGS server, KDC, schema master, and LDAP server). AFAIK for that kind of heterogeneous environment kerberos is the bast solution , it's secure, you can kerberized a lot of services (NFS etc.) Storing the information using LDAP is a great solution, but the way that Microsoft do it is not the simple way and not the standard way. - doron Jonathan Ben Avraham wrote: Hi Linux-IL members, I am considering setting up a heterogenous work environment with about 100 high-end Linux work stations, 40 MS Windows, and 10 Mac's. The underlying common authentication system will likely be LDAP. Would NIS or Active Directories be more appropriate for this type of environment? TIA, - yba = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Asterisk won't detect the line was hung up
Tzafrir Cohen wrote: You don't really need to set busypattern. (and is the busypattern here 500,500?) From the log files: [Dec 27 12:38:41] DEBUG[21319]: dsp.c:1360 ast_dsp_busydetect: ast_dsp_busydetect detected busy, avgtone: 520, avgsilence 500 Right. Not supported here. We have to resort to busydetect. Might be a good idea to add that to the FAQs about installing Asterisk in Israel. I never saw it mentioned. You have a typo there: $ grep asrecieved /usr/lib/asterisk/modules/chan_zap.so $ strings /usr/lib/asterisk/modules/chan_zap.so | grep asre asreceived That is a typo I have made too often. Ok. Is there any way to check whether it worked without without being present there? Just connecting to the machine via ssh and running the console does not reveal the caller ID number, but maybe I don't have the right debugging logs set up. Try: egrep '^(group|signalling|channel)' /etc/asterisk/zapata.conf You have your FXS channel (1) in group 0 and your FXO channels (7,8) in group 2. Try Zap/g2 instead of Zap/g1 . If I remember correctly, that's what I tried (backed by the commented out config I have). I'm not there right now to check again (I upgraded the software since last I tried). I'll post again when I know more. Shachar = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Asterisk won't detect the line was hung up
On Thu, Dec 27, 2007 at 05:43:33AM -0500, Tzafrir Cohen wrote: $ grep asrecieved /usr/lib/asterisk/modules/chan_zap.so $ strings /usr/lib/asterisk/modules/chan_zap.so | grep asre asreceived It's I before E Except After C http://alt-usage-english.org/excerpts/fxibefor.html HTH, Muli = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Signal handler priority
Ori Idan wrote: I have a process that creates several threads all of them in real time priorities. The process also has a signal handler, does anyone have any idea in what priority the signal handler will run? The same priority of the thread that happens to handle that signal. Please note that threads and signals mix very badly. You are highly advised to mask the signal in all threads and spawn a special thread just to handle the signal when it arrives, using sigwait(). Gilad -- Gilad Ben-Yossef [EMAIL PROTECTED] Codefidence. A name you can trust(tm) Web: http://codefidence.com | Cel: +972.52.8260388 IL: +972.3.7515563 ext. 201 | Fax:+972.3.7515503 = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Find process id of background ssh?
On Wed, 26 Dec 2007 17:50:44 +0200, Oded Arbel wrote: On Wed, 2007-12-26 at 16:02 +0200, Tom Rosenfeld wrote: pgrep ssh |tail -1 they way I read it ... ... it looks like there will be a race condition here. Maybe I could filter on session id or process group id if I would know how to set it. Here is a little trick that will solve the race condition - create a unique name for the ssh you run. Example of use in bash: SSH=`which ssh`# locate ssh on PATH NSSH=my_ssh_$$ # unique ssh name (you can add $USER, time ...) TSSH=/tmp/$NSSH# abs name rm -f $TSSH# may be we have leftovers ln -s $SSH $TSSH # create it on /tmp $TSSH command parameters # run the ssh in BG rm -f $TSSH# remove symbolic link (not needed any more) PID=`pgrep $NSSH` # find pid of MY ssh Ehud. -- Ehud Karni Tel: +972-3-7966-561 /\ Mivtach - Simon Fax: +972-3-7966-667 \ / ASCII Ribbon Campaign Insurance agencies (USA) voice mail and X Against HTML Mail http://www.mvs.co.il FAX: 1-815-5509341 / \ GnuPG: 98EA398D http://www.keyserver.net/Better Safe Than Sorry = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Fonts problem on Centos.
Hi all, We have recently upgraded to CentOS release 5. We are running application that was compiled with gcc 3.4 and Uses UI library named Ilog. In general the application run smoothly. The problem is that when we try to install Japanese fonts efont-unicode-bdf-0.4.2 the text in the window is not readable. This phenomena also happens when we run the application on Centos and redirect the display to our old redhat 9.0 display (In that case the fonts are installed on the RH machine). As mentioned the phenomena does not happen if we run the application on Redhat9 workstation. Any idea? Thanks in advace Yaron Kahanovitch = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Signal handler priority
I have a process that creates several threads all of them in real time priorities. The process also has a signal handler, does anyone have any idea in what priority the signal handler will run? -- Ori Idan
[OT] Looking for full-time Linux Java engineers
Hi, Thought this might be of interest to some of the members of this list. If interested, please reply to the mail listed below. Cheers, Rony We are seeking excellent s/w engineers to join GED-I Ltd in the development of s/w for Storage Data Security. This is a great opportunity for highly motivated engineers to join a start-up company in its early stages, make a crucial contribution to the company's success, and be rewarded accordingly. Note that these positions are for full-time employees, not freelancers or contractors. Offices are in the Netanya area. 1. Linux Kernel programmer An experienced kernel programmer for implementing security-related protocols in the kernel - Good understanding of Linux internals, especially networking - Experience in kernel programming - Strong C coding skills 2. Linux C/C++ programmer An experienced C/C++ programmer for writing security related applications under Linux. Requirements: - At least four years industrial experience writing C and C++ applications under Linux (not Windows Visual Studio). - Experience with performance optimizations - Familiarity with cryptography and data security - an advantage 3. A Web Jaba programmer with 2 years experience, preferrably familiar with Spring and JSF. Mail CV to: [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Does anyone maintain http://www.ivrix.org.il/ nowadays?
Today I tried to access it, and it took long time until my Web browser reported that it has a problem. --- Omer P.S.: I am looking for a Free Hebrew-English and/or English-Hebrew dictionary file. The nearest one, which I found is http://www.freelang.net/dictionary/hebrew.html but the Hebrew words are transcribed in Latin letters rather than in Hebrew letters. -- Eli Marmor's Law: netiquette forbids people from replying I agree with you to messages in mailing lists, skewing discussions in favor of the troublemaker type of dissenters. My own blog is at http://www.zak.co.il/tddpirate/ My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with which I may be affiliated in any way. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Fonts problem on Centos.
Hi Yaron, I'm not sure, it's just a shot in the dark, but could you check the DPI settings on a machine that shows the font well? If you don't have such a machine, try to play with the DPI settings to see if you get better results.. Also, you might want to check if antialiasing is on or off and if this font supports hinting for antialiasing.. Thanks, Hetz On Dec 27, 2007 4:54 PM, [EMAIL PROTECTED] wrote: Hi all, We have recently upgraded to CentOS release 5. We are running application that was compiled with gcc 3.4 and Uses UI library named Ilog. In general the application run smoothly. The problem is that when we try to install Japanese fonts efont-unicode-bdf-0.4.2 the text in the window is not readable. This phenomena also happens when we run the application on Centos and redirect the display to our old redhat 9.0 display (In that case the fonts are installed on the RH machine). As mentioned the phenomena does not happen if we run the application on Redhat9 workstation. Any idea? Thanks in advace Yaron Kahanovitch = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] -- Skepticism is the lazy person's default position. my blog (hebrew): http://benhamo.org = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Run out of inodes
Biran, Yahav (Yahav) wrote:
df -I show:
[EMAIL PROTECTED] mpower]# df -i
FilesystemInodes IUsed IFree IUse% Mounted on
/dev/cciss/c0d0p6 262144 10442 2517024% /
/dev/cciss/c0d0p1 26104 54 260501% /boot
none 1013170 1 10131691% /dev/shm
/dev/cciss/c0d0p23989888 616728 3373160 16% /export/home
/dev/cciss/c0d0p7 131616 20 1315961% /tmp
/dev/cciss/c0d0p33842720 127472 37152484% /usr
/dev/cciss/c0d0p8 131616 131616 0 100% /var
Therefore I can't run any rpm nor up2date.
How one can clean up some inodes?
The usual case is a directory which has a lot of files (usually small,
otherwise the FS will hit in size limits). An example might be empty log
files.
The best way to find problematic directory is to look for a large
directory. Don't mix between the directory size itself, and the size of
it's contents.
a small example to clarify:
$ ls -ld /tmp/inode-test/
drwxr-xr-x 2 kaplan kaplan 1024 2007-12-28 02:20 /tmp/inode-test/
$ for i in {1..5000}; do
touch /tmp/inode-test/$i
done
$ ls -ld /tmp/inode-test/
drwxr-xr-x 2 kaplan kaplan 60416 2007-12-28 02:29 /tmp/inode-test/
Notice the directory size, while each of the files in it is 0 in size.
From here, I'm sure you'll manage to find the large directory by yourself.
Good luck.
--
Lior Kaplan
[EMAIL PROTECTED]
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
