The forum will experience an outage sometime between February 10 at 7:00pm EST and February 12 at 11:59 EST for installation of security updates. The actual time and duration of the outage are unknown but attempts will be made to minimize the downtime. We apologize for any inconvenience.

Does anyone have experience of this assertion and what might be causing it?

It happens fairly regularly with a particular database, but only during a backup.

The database passes both express and full validation. We have also tried restoring a backup that pre-dates the first occurrence (then applying subsequent logs).

The only reference to it I can find is in the old NTTP newsgroups, where it was suggested that it might be connected to a VMWare bug - but in this case no virtualization is involved.

We are doing a full unload / reload just in case it really is a database corruption - but given that restoring the backup didn't fix it, the validation passes and the other case - it seems increasing unlikely that that is the answer. This will take a while to schedule as the database is about 80GB and is used 24x7.

asked 07 Oct '13, 06:44

Justin%20Willey's gravatar image

Justin Willey
6.4k101132197
accept rate: 20%

edited 07 Oct '13, 06:45

1

Only two google hits https://www.google.com/#q=%22invalid+access+to+a+null+image%22

...one unavailable right now (probably the one you found), the other on a Cisco forum says "corrupt database". As such, it's pretty rare, so you probably won't find anyone with experience on this forum. Try calling tech support because they have a vast internal problem report database.

(07 Oct '13, 07:51) Breck Carter

"Invalid Access to a NULL image" means that we failed to allocate a cache page when auto-growing the cache - the memory page received back was NULL when checking it, and we assert.

We have seen this from a few customers and yes, most of those have been associated with virtualized environments. I don't think we've yet had a reproducible in-house of the behaviour to diagnose the memory conditions precisely.

Most of our customers were able to remotely work-around the issue by strictly setting the "-ch" parameter on the server-start up switch to an explicit value - does that also work around the behaviour for yourself?

permanent link

answered 07 Oct '13, 10:24

Jeff%20Albion's gravatar image

Jeff Albion
10.7k171174
accept rate: 24%

Thanks Jeff - that is really useful. I will try that. This is a dedicated server so we could try a fixed cache - would that address the same issue?

(07 Oct '13, 10:40) Justin Willey
Replies hidden
2

To the best of our understanding, yes, turning off auto-sizing of the cache and fixing the cache size should also work here. Let us know if that works for you.

(07 Oct '13, 10:58) Jeff Albion

Thanks again - I'll let you know how we get on.

(07 Oct '13, 12:22) Justin Willey
2

Fixing the cache size seems to have done the trick - no assertions since then. - Many thanks.

(for anyone else coming across this - we used -c to set the initial cache size and -ca 0 to prevent re-sizing)

(11 Oct '13, 12:57) Justin Willey
Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:

×31
×9

question asked: 07 Oct '13, 06:44

question was seen: 2,861 times

last updated: 11 Oct '13, 12:59