[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: cvs rtag: cannot exec ...: Argument list too long

From: Mark E. Hamilton
Subject: Re: cvs rtag: cannot exec ...: Argument list too long
Date: Fri, 02 Dec 2005 11:47:54 -0700
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913

Jim Searle wrote:
For a directory with a large number of files I get this error:
cvs tag: cannot exec .../ Argument list too long

I searched the mailing lists and find references of this issue that are very
old, but I'm not able to find any fix...

Any help would be appreciated.
- Jim


This is an operating system problem. Some internal buffer is to small for the number of files in the directory. I've seen it in on a number of different types of systems (IBM, SGI, Linux.) On most any system you reproduce this error with:

ls `find /some/big/directory/hierarchy`

(You can back up the hierarchy untill you go over the limit if you want to find out about what it is.)

You didn't provide enough context, so I'm guessing that this is happening in a script invoked by taginfo. From the man page the command would look something like this:

.../ tagname operation repository {filename revision} ...

If a script (instead of does this:

echo $@

the output looks something like this:

sahp6613% cvs rtag test_tag2 dir1
test_tag2 add /var/scratch/mhamilt/cvsrepos/test/dir1 test.dat 1.1 test1.dat 1.1

Basically, you've just got too many {filename revision} pairs on the command line, which means that you've got too many files in that repository directory. (I suspect that if you went into that repository directory and did an 'ls *' you'd get the same error, though depending on how close to the limit you are the revision numbers for each file could be what's pushing you over.)

The two solutions that I know of are to find out where this buffer is configured (in the kernal?) and make it bigger, or reduce the number of files you've got in that repository directory.

Mark E. Hamilton
Orion International Technologies, Inc.
Sandia National Laboratory, NM.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]