Uploaded image for project: 'RHEL'
  1. RHEL
  2. RHEL-12009

The `sar -B` reports '%vmeff' beyond 100

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Won't Do
    • Icon: Normal Normal
    • None
    • rhel-9.0.0
    • sysstat
    • None
    • Low
    • rhel-sst-cs-plumbers
    • ssg_core_services
    • 5
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • Known Issue
    • Hide
      .The `%vmeff` metric from the `sysstat` package displays incorrect values

      The `sysstat` package provides the `%vmeff` metric to measure the page reclaim efficiency. The values of the `%vmeff` column returned by the `sar -B` command are incorrect because `sysstat` does not parse all relevant `/proc/vmstat` values provided by later kernel versions. To work around this problem, you can calculate the `%vmeff` value manually from the `/proc/vmstat` file. For details, see link:https://access.redhat.com/solutions/7027076[Why the `sar(1)` tool reports `%vmeff` values beyond 100 % in RHEL 8 and RHEL 9?]
      Show
      .The `%vmeff` metric from the `sysstat` package displays incorrect values The `sysstat` package provides the `%vmeff` metric to measure the page reclaim efficiency. The values of the `%vmeff` column returned by the `sar -B` command are incorrect because `sysstat` does not parse all relevant `/proc/vmstat` values provided by later kernel versions. To work around this problem, you can calculate the `%vmeff` value manually from the `/proc/vmstat` file. For details, see link: https://access.redhat.com/solutions/7027076 [Why the `sar(1)` tool reports `%vmeff` values beyond 100 % in RHEL 8 and RHEL 9?]
    • Done
    • None

      Description of problem:
      The `sar -B` commands report the paging statistics where the output
      includes '%vmeff' column.
      The `%vmeff` as per the man page is [pgsteal / pgscan].

      Version-Release number of selected component (if applicable):
      sysstat-12.5.4-5.el9.x86_64

      How reproducible:
      Always

      Steps to Reproduce:
      1. Stress the system for memory utilization so that the cache is reclaimed & swapping occurs.
      2. sar -B
      3. And look for the `%vmeff` column.

      Actual result:

      1. sar -B 1
        Linux 5.14.0-284.25.1.el9_2.x86_64 (9rhel) 08/09/2023 x86_64 (1 CPU)

      04:36:26 PM pgpgin/s pgpgout/s fault/s majflt/s pgfree/s pgscank/s pgscand/s pgsteal/s %vmeff
      04:36:27 PM 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
      04:36:28 PM 0.00 0.00 0.00 0.00 4.00 0.00 0.00 0.00 0.00
      04:36:29 PM 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.00
      04:36:30 PM 491.09 105586.14 60935.64 4.95 57935.64 239636.63 9034.65 94788.12 38.12
      04:36:31 PM 12310.00 538256.00 80468.00 133.00 92348.00 204523.00 43644.00 164542.00 66.30
      04:36:32 PM 1031.67 167920.00 74993.75 13.33 79612.08 172162.08 15215.00 149747.50 79.92
      04:36:34 PM 994.29 499363.81 122150.48 19.05 138315.24 319519.05 70208.57 251845.71 64.62
      04:36:35 PM 13865.26 254842.11 21938.95 441.05 48472.63 131057.89 11048.42 81572.63 57.40
      04:36:36 PM 153948.62 129625.69 43133.03 4892.66 45727.52 109327.52 0.00 69844.04 63.89
      04:36:37 PM 72123.08 9.89 19817.58 2248.35 17915.38 10175.82 0.00 20351.65 200.00 <<-----
      04:36:38 PM 29152.00 0.00 8227.00 919.00 7413.00 7411.00 0.00 14806.00 199.78 <<-----
      04:36:39 PM 105044.25 205497.35 29342.48 3339.82 37755.75 151597.35 0.00 62493.81 41.22
      04:36:40 PM 107365.52 0.00 29221.84 3340.23 19342.53 19228.74 0.00 38340.23 199.39 <<-----
      04:36:41 PM 399295.05 348940.59 111702.97 12735.64 106972.28 246384.16 0.00 200459.41 81.36
      04:36:42 PM 322668.00 339464.00 89742.00 10247.00 93473.00 273802.00 0.00 170922.00 62.43
      04:36:43 PM 408888.00 380136.00 113442.00 12926.00 105087.00 276010.00 0.00 194792.00 70.57
      04:36:44 PM 390824.00 389380.00 108543.00 12304.00 107732.00 311288.00 0.00 198284.00 63.70
      04:36:46 PM 109600.00 248129.17 30189.06 3431.77 45426.56 148714.06 0.00 76365.62 51.35
      04:36:47 PM 70936.00 0.00 19666.00 2239.00 31.00 0.00 0.00 44.00 0.00
      04:36:48 PM 108696.00 0.00 30310.00 3477.00 24087.00 24083.00 0.00 48148.00 199.93 <<-----

      Expected result:
      The value of %vmeff should be <=100%

      Additional info:
      The Bugzilla has already been raised for RHEL 8[1] and it was identified that the changes in the /proc/vmstat fields caused this issue. The issue has been raised to upstream[2].

      Raising this Bugzilla to track for RHEL9.

      [1] https://bugzilla.redhat.com/show_bug.cgi?id=2228949
      [2] https://github.com/sysstat/sysstat/issues/343#issuecomment-1667640798

              rh-ee-lzaoral Lukáš Zaoral
              rhn-support-aquadri Abdul Raheman Quadri
              Lukáš Zaoral Lukáš Zaoral
              qe-baseos-daemons qe-baseos-daemons
              Mugdha Soni Mugdha Soni
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Created:
                Updated:
                Resolved: