Sponsored Content
Top Forums Shell Programming and Scripting Unable to catch the redirection error when the disk is full Post 302804623 by Pruthviraj_shiv on Thursday 9th of May 2013 03:13:37 AM
Old 05-09-2013
Hi Scru,
Yes done that too , but even sed created an incomplete file.

I have other options like "taking the count of the resultant file & comparing with original file". needless to say this is a huge overhead on bigfiles , when we have 100's of jobs running parallely.
 

8 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

disk full

Please solve the following NOTICE HTFS:No space on dev hd(1/42) (2 Replies)
Discussion started by: msuheel
2 Replies

2. AIX

AIX 5.3 errpt full of message: DISK OPERATION ERROR

Hi All, Can anyone explain me the meanning of the following errors: LABEL: SC_DISK_ERR2 IDENTIFIER: B6267342 Description DISK OPERATION ERROR Probable Causes DASD DEVICE Failure Causes DISK DRIVE DISK DRIVE ELECTRONICS Recommended Actions PERFORM PROBLEM DETERMINATION... (1 Reply)
Discussion started by: gianlu
1 Replies

3. Solaris

Unable to bring a disk from error to online state.

Hello experts. I am using Solaris 10(2005) on intel machine. I have installed Veritas Volume manager 5.0. I am unable to bring a disk error to online state. I would like to bring that disk to CDS format. Commands i used and output are.... #vxdisk list DEVICE TYPE ... (11 Replies)
Discussion started by: younus_syed
11 Replies

4. Linux

Unable to remove file using rm: Disk space is full

Hi all, My disk space is 100% full. df -k <dir> -> 100% One of my debug files consume huge amount of space and i want to remove the same to start off fresh debugs. However i'm unable to remove the file giving out the following error message: rm -f debug.out22621 rm: cannot remove... (8 Replies)
Discussion started by: Pankajakshan
8 Replies

5. Linux

Disk full 100%

one of my servers / was full by 100% i cleard some space, now though i have enough space on / partition still df is showing disk usage as 100% am not able to create any single txt file ? why so ? (3 Replies)
Discussion started by: bryanabhay
3 Replies

6. Shell Programming and Scripting

Unable to catch the output after core dump and bus error

I have a weird situation in which the binary dumps core and gives bus error. But before dumping the core and throwing the buss error, it gives some output. unfortunately I can't grep the output before core dump db2bfd -b test.bnd maxSect 15 Bus Error (core dumped) But if I do ... (4 Replies)
Discussion started by: rakeshou
4 Replies

7. HP-UX

Unable to get full FS space after mounting

Hi, I am unable to get the full FS space, as /home is 100% utilized and after deleting unwanted files, its still 100%. After checking the du -sk * | sort -n output and converting it to MBs, the total sizes comes out to be 351 MBs only however the lvol is of 3GB. I don't know where is all the space... (2 Replies)
Discussion started by: Kits
2 Replies

8. Shell Programming and Scripting

Disk full alerts

i want to create 1 script to monitor 1 particular filesystem out of the diferent filesystems. if disk space of that particular filesystem increases by 80% it sends an alert mail to an email id ---------- Post updated at 04:18 PM ---------- Previous update was at 04:17 PM ---------- no. I am... (1 Reply)
Discussion started by: rakeshhhhhhhh
1 Replies
iopending(1m)							   USER COMMANDS						     iopending(1m)

NAME
iopending - plot number of pending disk events. Uses DTrace. SYNOPSIS
iopending [-c] [-d device] [-f filename] [-m mount_point] [interval [count]] DESCRIPTION
This samples the number of disk events that are pending and plots a distribution graph. By doing this the "serialness" or "parallelness" of disk behaviour can be distinguished. A high occurance of a pending value of more than 1 is an indication of saturation. Since this uses DTrace, only users with root privileges can run this command. OPTIONS
-c clear screen -d device instance name to snoop (eg, dad0) -f filename full pathname of file to snoop -m mount_point mountpoint for filesystem to snoop EXAMPLES
Default output, print I/O summary every 1 second, # iopending Print 10 second samples, # iopending 10 Print 12 x 5 second samples, # iopending 5 12 Snoop events on the root filesystem only, # iopending -m / FIELDS
value number of pending events, 0 == idle count number of samples @ 1000 Hz load 1 min load average disk_r total disk read Kb for sample disk_w total disk write Kb for sample IDEA
Dr Rex di Bona DOCUMENTATION
See the DTraceToolkit for further documentation under the Docs directory. The DTraceToolkit docs may include full worked examples with ver- bose descriptions explaining the output. EXIT
iopending will run forever until Ctrl-C is hit, or the specified count is reached. AUTHOR
Brendan Gregg [Sydney, Australia] SEE ALSO
iosnoop(1M), iotop(1M), dtrace(1M) version 0.60 Nov 01, 2005 iopending(1m)
All times are GMT -4. The time now is 08:19 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy