That sounds like a Java heap allocation. IS this running WebSphere? If your GC 
doesn't run for 49 days something is Very Wrong. It should be running
periodically even in an idle WebSphere region. I recall seeing something like 
this when a JVM was sized too small for the number of applications that
were deployed. Increasing the low water mark to the average heap size and 
adding some headroom to the maximum possible value alleviated this issue.

In WebSphere the min/max heap values are there to allow the JVM heap to grow in 
response to demand. If your min/max values are too close or are even
the same, your JVM is already as large as it is allowed to get and there was an 
allocation that could not be satisfied before the GC action took
place. I suspect that after the allocation would have worked after GC 
processing.



             Rick Barlow <[EMAIL PROTECTED]>
             Sent by: Linux on 390 Port
             <[email protected]>                                          
                                                                   To
                                                                     
[email protected]
                                                                                
                                                                   cc
             05/17/2007 06:11 AM
                                                                                
                                                              Subject
                                                                     need help 
to understand Allocation Failure message
                            Please respond to
               Linux on 390 Port <[email protected]>








We had an occurrence where a group of Linux guests produced Allocation
Failure messages which include a time value of about 49 days "since last
AF".  A sample is included below.  I would like to know where I can find
information to explain that message.  Does the AF in "since last AF"
indicate time since the last allocation failure?  Our middleware support
seems to think that it indicates the time since the last GC and they have
never seen a time value that large.  They are concerned that the system
clock was somehow adjusted but there is no indication that the clock was
reset.  I would appreciate and suggestions about what to look at.

<AF[275]: Allocation Failure. need 520 bytes, 4290762359 ms since last AF>
<AF[275]: managing allocation failure, action=1 (0/965512360)
(23823992/25521496)>
<GC(275): GC cycle started Sat May 12 08:14:08 2007

____________________________________
Rick Barlow
Systems Engineering Consultant
Nationwide Services Co., Technology Solutions
Mainframe, z/VM and System z Linux Support
One Nationwide Plaza  3-20-13
Columbus OH 43215-2220   U.S.A
Voice: (614) 249-5213    Fax: (614) 677-0821
mailto:[EMAIL PROTECTED]

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to