Hello,
with a customer of ours, that have some quite large blob spaces we get a timeout warning, when starting a "onstat -d" on Informix7.31/HPUX11:
Chunks
address chk/dbs offset size free bpages flags pathname
The server did not respond within 21 seconds. Timing out.
Warning: blob chunk page usage statistics may be inaccurate.
caf57218 1 1 0 750000 748625 PO- /dev/dbspaces/h1rootdbs
caf57620 2 2 0 100000 947 PO- /dev/dbspaces/h1llogs
caf57728 3 3 0 100000 947 PO- /dev/dbspaces/h1plogs
caf57830 4 4 0 1024000 ~20478 20480 POB caf588b0 20 16 0 1024000 523947 PO- /dev/dbspaces/h1ifounter01
caf589b8 21 17 0 1024000 ~0 1024000 POB /dev/dbspaces/h1ifounterblob01
caf58ac0 22 17 0 1024000 ~0 1024000 POB /dev/dbspaces/h1ifounterblob02
caf58bc8 23 17 0 1024000 ~0 1024000 POB /dev/dbspaces/h1ifounterblob03
caf58cd0 24 17 0 1024000 ~627760 1024000 POB /dev/dbspaces/h1ifounterblob04
caf58dd8 25 17 0 1024000 ~1019951 1024000 POB /dev/dbspaces/h1ifounterblob05
caf58ee0 26 17 0 1024000 ~1019951 1024000 POB /dev/dbspaces/h1ifounterblob06

Has anyone experienced that too?
What does it mean? Should we do anything about that?

Thanks, Peter

Normally when I get these type of messages with onstat commands:

"The server did not respond within 21 seconds. Timing out."

I have to bounce the engine. Do any of the other onstat commands time out?

I believe this is to do with the onstat commands needing access to critical memory structures that are changing when the command in in play. Similar to the old 'changing memory structure caused command-line termination' errors onstat used to give to busy Solaris systems.

I asked someone I know, who administrates a few larger INFORMIX DBSs, who said:

Should be no problem, occurs alway when there is more than a certain number of blob chunks.
It´s only a warning. The Server can not measure the exact usage of blob space in time. For exact values use onmonitor.