[Info-vax] BL860i2 blade unexplained system disk slowdowns

[chris]

On 06/10/22 14:21, Kenneth Randell wrote:
> On Thursday, June 9, 2022 at 9:43:08 PM UTC-4, Edgar Ulloa wrote:
>> Hi Kenneth
>>
>> Inside the onboard administrator do you see something abnormal in that blade?
>>
>> Show interconnect info all
>>
>> Show health
>>
>> Already at the operating system level, is there a process that is writing a lot?
>>
>> $monitor proc/topcpu/int=3
>> $monitor proc/topfault
>> $monitor dlock
>> $monitor rms
>>
>> System disk space?
>>
>> What does swapfile and pagefile look like?
>>
>> Has any file on the entire system disk reached its version limit?
>>
>> Any error during boot?
>> $type sys$system:startup.log
>>
>> And how about the network, any collision?
>> tcpip>netstat –i
>> Lancp>sh dev/count ewa0 (your card)
>>
>> if the account file is defined and exist?
>> $sh log acco
>> $dir $1$xxx:[xxxxxxx]acc*.dat (same place that log show)
>>
>> some process in rwast state... example opcom?
>> I look forward to your comments
>
> There are no indications of any issues from the O/A, ILO, or any of the above VMS commands.  The best indicator I have is the disk response time as reported by the SDA XFC extension, but that still does not explain 'why'.
>
> This value for I/O response time is 'normal':
>
> I/O Response Times (This Volume)
>    Overall Average:                0.3301 milliseconds
>    Cache Hit:                      0.0158 milliseconds
>    Disk:                           4.6630 milliseconds
>    Accuracy:                     99 %
>
> This value is from earlier today when the system became non-reponsive.
>
> I/O Response Times (This Volume)
>    Overall Average:                4.7281 milliseconds
>    Cache Hit:                      0.0172 milliseconds
>    Disk:                          80.7186 milliseconds
>    Accuracy:                    100 %
>

Sounds like it might be a hard disk problem. Something failing
as it warms up. Try replacing it seems like the obvious first
step...

Chris