Discussion:
[Enterprise Guide] An I/O error has occurred on file
(too old to reply)
Sandro Saitta
2009-05-14 14:42:51 UTC
Permalink
Hello,

After running my code, I obtain this error message:

"ERROR: An I/O error has occurred on file xxxxx"

It is says on SAS website that it could be due to disc space
limitation. I have 500Gb so it should not be a problem.

Does anyone have an idea for such an error?

Thanks.
Sandro.
Gerhard Hellriegel
2009-05-14 17:56:53 UTC
Permalink
another idea:
the IO errors often come from disk-full situations, but not always. It is
simply a unexpected end of file or something that differs between the
header and the physical file. End-of-disk-space simply cuts a file at a
unexpected location. There might be other reasons for that, e.g. a
unexpected task-stop (killed?) or some other errors. Not "normal" SAS
errors, like "missing semicolon" or such things, but something which could
not be handled by the error routines, primarily a external reason (perhaps
a power breakdown?)
Like Miiiikeee wrote: sometimes a rerun of the job which buids the dataset
helps. If you permanently get that error, always at the same place and you
can excluide all the external problems (also a network problem can be a
reason, if you write to a remote disk!), we should continue searching.

By the way: on a mainframe I had once the thing that there was REALLY a
problem with a disk which caused a IO error! After several trials with
several runs with the problem and several without, I gave up and informed
the system people. They checked the disk and took it out of the pool and
the problem disappeared and reappeared again next day. The day after, they
replaced a complete EMC box, where the virtual disk was on. They had a
kind of controler problem, not on the disks... So in very rare
circumstances also a real (virtual or physical) IO problem outside of SAS
can lead to that problems. If you are under win and your disk is local,
did you check it with chkdsk (only the last thing, because I don't think
that this is the reason in that case)?

Gerhard



On Thu, 14 May 2009 13:29:56 -0400, Michael Raithel
Dear SAS-L-ers,
Post by Sandro Saitta
"ERROR: An I/O error has occurred on file xxxxx"
It is says on SAS website that it could be due to disc space
limitation. I have 500Gb so it should not be a problem.
Does anyone have an idea for such an error?
Sandro, knowing more about your hardware config and OS would probably net
1. Do you really have 500Gb free for building your SAS file? If that is
a local hard drive, or exclusively yours, you might want to check it to
see if there is enough room on it to build the (apparently large) SAS file
that you are attempting to build.
2. If you are running on a shared server (UNIX, Linux, Windows server,
etc.) perhaps other SAS programmers were running programs that soaked up a
lot of space too. Your program bumped into that dynamic space limitation
and failed. Then, their programs completed. You look out there and see
500Gb of free space and exclaim: WTH!
3. Are you sure that your erroneous file is being built on the 500Gb
drive?
4. Perhaps the I/O subsystem burped and threw a data write error that SAS
couldn't handle. Have you tried to rerun your SAS program--and if so, did
it poop-out in the same spot?
5. If you are thoroughly flummoxed by this error again and again and
again; try breaking your output SAS data set up and writing on two
different disk drives.
These are always interesting types of problems; so I would love to see
your final solution of, or resolution to, this issue.
Sandro, best of luck in all of your SAS endeavors!
I hope that this suggestion proves helpful now, and in the future!
Of course, all of these opinions and insights are my own, and do not
reflect those of my organization or my associates. All SAS code and/or
methodologies specified in this posting are for illustrative purposes only
and no warranty is stated or implied as to their accuracy or
applicability. People deciding to use information in this posting do so at
their own risk.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Michael A. Raithel
"The man who wrote the book on performance"
Author: Tuning SAS Applications in the MVS Environment
Author: Tuning SAS Applications in the OS/390 and z/OS Environments,
Second Edition
http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172
Author: The Complete Guide to SAS Indexes
http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Men are more satirical from vanity than from malice. - Francois de La
Rochefoucauld
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Sandro
2009-05-15 08:44:12 UTC
Permalink
Hi,

Thanks all for your answers. In fact we found the problem. By changing
all the paths from:

\\server_name\my_folder\my_file.sas

to:

k:\my_folder\my_file.sas

where k:\ is the mapping from the server, then it works!

Thanks.
Sandro.
Post by Gerhard Hellriegel
the IO errors often come from disk-full situations, but not always. It is
simply a unexpected end of file or something that differs between the
header and the physical file. End-of-disk-space simply cuts a file at a
unexpected location. There might be other reasons for that, e.g. a
unexpected task-stop (killed?) or some other errors. Not "normal" SAS
errors, like "missing semicolon" or such things, but something which could
not be handled by theerrorroutines, primarily a external reason (perhaps
a power breakdown?)Like Miiiikeee wrote: sometimes a rerun of the job which buids the dataset
helps. If you permanently get thaterror, always at the same place and you
can excluide all the external problems (also a network problem can be a
reason, if you write to a remote disk!), we should continue searching.
By the way: on a mainframe I had once the thing that there was REALLY a
problem with a disk which caused a IOerror! After several trials with
several runs with the problem and several without, I gave up and informed
the system people. They checked the disk and took it out of the pool and
the problem disappeared and reappeared again next day. The day after, they
replaced a complete EMC box, where the virtual disk was on. They had a
kind of controler problem, not on the disks... So in very rare
circumstances also a real (virtual or physical) IO problem outside of SAS
can lead to that problems. If you are under win and your disk is local,
did you check it with chkdsk (only the last thing, because I don't think
that this is the reason in that case)?
Gerhard
On Thu, 14 May 2009 13:29:56 -0400, Michael Raithel
Dear SAS-L-ers,
"ERROR: AnI/Oerrorhas occurred on file xxxxx"
It is says on SAS website that it could be due to disc space
limitation. I have 500Gb so it should not be a problem.
Does anyone have an idea for such anerror?
Sandro, knowing more about your hardware config and OS would probably net
1. Do you really have 500Gb free for building your SAS file?  If that is
a local hard drive, or exclusively yours, you might want to check it to
see if there is enough room on it to build the (apparently large) SAS file
that you are attempting to build.
2. If you are running on a shared server (UNIX, Linux, Windows server,
etc.) perhaps other SAS programmers were running programs that soaked up a
lot of space too.  Your program bumped into that dynamic space limitation
and failed.  Then, their programs completed.  You look out there and see
500Gb of free space and exclaim:  WTH!
3. Are you sure that your erroneous file is being built on the 500Gb
drive?
4. Perhaps theI/Osubsystem burped and threw a data writeerrorthat SAS
couldn't handle.  Have you tried to rerun your SAS program--and if so, did
it poop-out in the same spot?
5. If you are thoroughly flummoxed by thiserroragain and again and
again; try breaking your output SAS data set up and writing on two
different disk drives.
These are always interesting types of problems; so I would love to see
your final solution of, or resolution to, this issue.
Sandro, best of luck in all of your SAS endeavors!
I hope that this suggestion proves helpful now, and in the future!
Of course, all of these opinions and insights are my own, and do not
reflect those of my organization or my associates. All SAS code and/or
methodologies specified in this posting are for illustrative purposes only
and no warranty is stated or implied as to their accuracy or
applicability. People deciding to use information in this posting do so at
their own risk.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Michael A. Raithel
"The man who wrote the book on performance"
Author: Tuning SAS Applications in the MVS Environment
Author: Tuning SAS Applications in the OS/390 and z/OS Environments,
Second Edition
http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172
Author: The Complete Guide to SAS Indexes
http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Men are more satirical from vanity than from malice. - Francois de La
Rochefoucauld
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++- Hide quoted text -
- Show quoted text -- Hide quoted text -
- Show quoted text -
Michael Raithel
2009-05-14 17:29:56 UTC
Permalink
Dear SAS-L-ers,
Post by Sandro Saitta
"ERROR: An I/O error has occurred on file xxxxx"
It is says on SAS website that it could be due to disc space
limitation. I have 500Gb so it should not be a problem.
Does anyone have an idea for such an error?
Sandro, knowing more about your hardware config and OS would probably net better guesses. Here are some off-the-cuff things to consider:

1. Do you really have 500Gb free for building your SAS file? If that is a local hard drive, or exclusively yours, you might want to check it to see if there is enough room on it to build the (apparently large) SAS file that you are attempting to build.

2. If you are running on a shared server (UNIX, Linux, Windows server, etc.) perhaps other SAS programmers were running programs that soaked up a lot of space too. Your program bumped into that dynamic space limitation and failed. Then, their programs completed. You look out there and see 500Gb of free space and exclaim: WTH!

3. Are you sure that your erroneous file is being built on the 500Gb drive?

4. Perhaps the I/O subsystem burped and threw a data write error that SAS couldn't handle. Have you tried to rerun your SAS program--and if so, did it poop-out in the same spot?

5. If you are thoroughly flummoxed by this error again and again and again; try breaking your output SAS data set up and writing on two different disk drives.

These are always interesting types of problems; so I would love to see your final solution of, or resolution to, this issue.

Sandro, best of luck in all of your SAS endeavors!


I hope that this suggestion proves helpful now, and in the future!

Of course, all of these opinions and insights are my own, and do not reflect those of my organization or my associates. All SAS code and/or methodologies specified in this posting are for illustrative purposes only and no warranty is stated or implied as to their accuracy or applicability. People deciding to use information in this posting do so at their own risk.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Michael A. Raithel
"The man who wrote the book on performance"
E-mail: ***@westat.com

Author: Tuning SAS Applications in the MVS Environment

Author: Tuning SAS Applications in the OS/390 and z/OS Environments, Second Edition

http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=58172

Author: The Complete Guide to SAS Indexes

http://www.sas.com/apps/pubscat/bookdetails.jsp?catid=1&pc=60409

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Men are more satirical from vanity than from malice. - Francois de La Rochefoucauld
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Continue reading on narkive:
Loading...