performancecopilot/xsos-pcp

pcp-xsos fails to fetch disk related data (using -d option) when run on a archive with -O option.

Closed this issue · 1 comments

Hello Team,

The tool fails to fetch the disk related data (using -d option) when run on a archive with -O option.

Archive details :
[Ani@rhel8 xsos-pcp]$ pcp -a /var/log/pcp/pmlogger/rhel8/20240606.0.xz
Performance Co-Pilot configuration on rhel8:
  archive: /var/log/pcp/pmlogger/rhel8/20240606.0.xz
 platform: Linux rhel8 4.18.0-513.24.1.el8_9.x86_64 #1 SMP Thu Mar 14 14:20:09 EDT 2024 x86_64
 hardware: 4 cpus, 1 disk, 1 node, 3653MB RAM
 timezone: EDT+4
 services: pmcd
     pmcd: Version 5.3.7-20, 9 agents, 4 clients
     pmda: root pmcd proc pmproxy xfs linux mmv kvm jbd2
 pmlogger: rhel8: /var/log/pcp/pmlogger/rhel8/20240606.00.58
     pmie: primary engine: /var/log/pcp/pmie/rhel8/pmie.log
[Ani@rhel8 xsos-pcp]$ pmdumplog -L -z /var/log/pcp/pmlogger/rhel8/20240606.0.xz
Note: timezone set to local timezone of host "rhel8" from archive
Log Label (Log Format Version 2)
Performance metrics from host rhel8
    commencing Thu Jun  6 00:58:30.625194 2024
    ending     Sun Jun  9 22:59:35.499956 2024
Archive timezone: EDT+4
PID for pmlogger: 3489

[Ani@rhel8 xsos-pcp]$ git pull
Already up to date.
[Ani@rhel8 xsos-pcp]$ pwd
/home/Ani/xsos-pcp
$ ./pcp-xsos -O 'Jun 8 10:00:00' -a /var/log/pcp/pmlogger/rhel8/20240606.0.xz -d
STORAGE
  Whole Disks from /proc/partitions:
     disks, totalling 0 GiB (0 TiB)
    - - - - - - - - - - - - - - - - - - - - -
    Disk     Size in GiB
    ----     -----------
  Filesystem usage from df:
    Filesystem     1K-blocks     Used Available Use% Mounted on
If we omit the -O, it shows the data :
[Ani@rhel8 xsos-pcp]$ ./pcp-xsos -a /var/log/pcp/pmlogger/rhel8/20240606.0.xz -d
STORAGE
  Whole Disks from /proc/partitions:
     disks, totalling 0 GiB (0 TiB)
    - - - - - - - - - - - - - - - - - - - - -
    Disk     Size in GiB
    ----     -----------
  Filesystem usage from df:
    Filesystem     1K-blocks     Used Available Use% Mounted on
    /dev/mapper/rhel-root       17811456                     69% /
    /dev/vda1        1038336                     30% /boot

Also "Whole Disks from /proc/partitions:" does not show any details in all cases. I am not sure if this is expected.

[Ani@rhel8 xsos-pcp]$ cat /proc/partitions 
major   minor  #blocks  name 
252        0   20971520 vda  
252        1    1048576 vda1  
252        2   19921920 vda2   
11         0    1048575 sr0  
253        0   17821696 dm-0  
253        1    2097152 dm-1

I think this is possibly the same issue as #12 from a different angle - resolve the -O option syntax (add '@' at the front) and it should improve. Still, there's an issue of robust failure to be handled in the code.