Expanding TXSeries Volumes - Windows Edition

Technote (troubleshooting)


Problem(Abstract)

The Toolkit server (SFS and PPC Gateway) crashes or fails to start because a logical volume which is used as a data volume or a log volume has run out of space.

Symptom

Error message ENC-log-0258 in the msg log for the server.


Cause

Bad programming.

Resolving the problem

The following is a demonstration of expanding an logical volume that is backed by a Windows file.

The region has to be shutdown and no other programs accessing the server. The server is started in administration mode
(cicssfs -a /.:/cics/sfs/<sfsname>).

The details for the commands are found in the TXSeries Information/Knowledge Centers.

Expanded the log with the following steps:

Step 1: Start the SFS in Administrative mode.
C:\>cicssfs -a /.:/cics/sfs/realbig
ERZ038217I/0176: Waiting for subsystem 'cicssfs.Srealbig' to become active
ERZ038182I/0180: Server '/.:/cics/sfs/realbig' started successfully.


Step 2: The current log size in pages is needed for reference.
tkadmin query lvol -server /.:/cics/sfs/<sfsname> <log_volumename>

C:\>tkadmin query lvol -server /.:/cics/sfs/realbig log_Srealbig
Information about logical volume log_Srealbig
All sizes and offsets are in pages. Page Size is: 4 Kbytes
size: 16320 chunkSize: 64
Backing physical volumes (only 'clean' and 'dirty' volumes are active):
log_Srealbig (clean)

The size is 16320.


Step 3: Show the physical files currently making up the physical log volume.
tkadmin query pvol -server /.:/cics/sfs/<sfsname> <log_volumename>

C:\>tkadmin query pvol -server /.:/cics/sfs/realbig log_Srealbig
Information about physical volume log_Srealbig
All sizes and offsets are in pages. Page Size is: 4 Kbytes
Mapped to logical volume log_Srealbig
chunk size: 64
number of regions: 1
region 0: disk: C:\VAR\log_Srealbig offset: 0 size: 16320
total size: 16320

One file is used to make the log volume.


Step 4: Allocate another file for the log volume. 67m is the default log size.
c:\opt\cics\etc\filevol c:\var\<file name> 67m

C:\>c:\opt\cics\etc\filevol c:\var\log_Srealbig2 67m
67m bytes is 70254592 bytes.
Set c:\var\log_Srealbig2 to 70254592 bytes.


Step 5: Add the new file to the physical log volume. May take a minute or two.
tkadmin init disk -server /.:/cics/sfs/<sfsname> <file name> :

C:\>tkadmin init disk -server /.:/cics/sfs/realbig c:\var\log_Srealbig2
Initialized disk partition c:\var\log_Srealbig2
disk size (in pages): 17151


Step 6: Add the additional file to the physical volume for the log.
tkadmin expand pvol <log_volumename> 1 <file name> 0 -server /.:/cics/sfs/<sfsname>

C:\>tkadmin expand pvol log_Srealbig 1 c:\var\log_Srealbig2 0 -server /.:/cics/sfs/realbig
(no output)


Step 7: Show the added file in the physical volume. The "total size" number will be needed later for in another command.
tkadmin query pvol -server /.:/cics/sfs/<sfsname> <log_volumename>

C:\>tkadmin query pvol -server /.:/cics/sfs/realbig log_Srealbig
Information about physical volume log_Srealbig
All sizes and offsets are in pages. Page Size is: 4 Kbytes
Mapped to logical volume log_Srealbig
chunk size: 64
number of regions: 2
region 0: disk: C:\VAR\log_Srealbig offset: 0 size: 16320
region 1: disk: c:\var\log_Srealbig2 offset: 0 size: 17088
total size: 33408


Step 8: Enlarge the logical logical volume for the log volume. The logical volume maps over the physical volume.
tkadmin expand lvol <log_volumename> <total size number> -server /.:/cics/sfs/<sfsname>

C:\>tkadmin expand lvol log_Srealbig 33408 -server /.:/cics/sfs/realbig
(no output)


Step 9: Show the details of the log volume. The "size" should equal the "total size" from the tkadmin query pvol command.
tkadmin query lvol -server /.:/cics/sfs/<sfsname> <log_volumename>

C:\>tkadmin query lvol -server /.:/cics/sfs/realbig log_Srealbig
Information about logical volume log_Srealbig
All sizes and offsets are in pages. Page Size is: 4 Kbytes
size: 33408 chunkSize: 64
Backing physical volumes (only 'clean' and 'dirty' volumes are active):


The SFS can now be stopped and restarted in regular mode. The SFS will recover the transactions in play at the time of the log going full. It may take a minute or two to Start up. If the SFS crashes again it is usually due to the log not being expanded large enough to handle all the transactions in the log. The server msg log will show compression and out of space messages when the log is being recovered. The log will have message ENC-log-0258 at the end if there is not enough space in the volume. If this happens the log will need expanded again. You may end up having to expand the log to twice it's size to have recovery to finish correctly.
There is a 2GB limit to the log file. The log can not be expanded past 2GB. If the log is at 2GB before expansion the SFS CAN NOT be recovered. The SFS will need restored from the latest backup. This is one of the reasons to keep the log at 100MB. When the SFS log is close to 1MB the SFS should be deleted and reallocated with a 100MB log.

Rate this page:

(0 users)Average rating

Add comments

Document information


More support for:

TXSeries
CICS

Software version:

6.1, 6.2, 7.1, 8.1

Operating system(s):

Windows

Reference #:

1616736

Modified date:

2014-09-11

Translate my page

Machine Translation

Content navigation