Bug 6832

Summary: disk_io value in database doesn't match sometime with the value in the task buildstats file
Product: [Build System, Metadata & Runtime] Toaster Reporter: Ionut Chisanovici <ionutx.chisanovici>
Component: toasterAssignee: Alexandru Damian <alexandru.damian>
Status: VERIFIED DUPLICATE QA Contact: Alexandru Roman <alexandru.costinx.roman>
Severity: normal    
Priority: Medium CC: alexandru.c.georgescu, alexandru.costinx.roman, alexandru.damian, belen.barros.pena, bluelightning, cristiana.voicu
Version: 1.7   
Target Milestone: 1.8 M4   
Hardware: x86   
OS: Multiple   
Whiteboard:
OS type for building Yocto: --- Type of Regression: New (Never tested)
Verified: Documentation change: No (bug/feature does not impact docs)
Attachments:
Description Flags
do_extra_symlinks_stat_file
none
toaster.sqlite none

Description Ionut Chisanovici 2014-10-14 09:03:01 UTC
Created attachment 2186 [details]
do_extra_symlinks_stat_file

Hi,

Ex: For task do_extra_symlinks for libgcc-initial-4.9.1-r0 package the disk_io in database doesn't match the value from the task buildstats file:

Database:

select * from orm_task where id=1984;
          build_id = 1
    work_directory = 
         cpu_usage = 0
     task_executed = 1
           disk_io = 0
       line_number = 0
path_to_sstate_obj = 
   sstate_checksum = b63d8944199267f46ff1e2aa84a77fcb
        source_url = 
             order = 1136
      elapsed_time = 0.02
           logfile = /home/ionut/work/weekly/poky/build/tmp/work/i586-poky-linux/libgcc-initial/4.9.1-r0/temp/log.do_extra_symlinks.3754
         recipe_id = 257
         task_name = do_extra_symlinks
           message = recipe libgcc-initial-4.9.1-r0: task do_extra_symlinks: Succeeded
           outcome = 0
                id = 1984
     sstate_result = 0
       script_type = 2

Stat file:

cat tmp/buildstats/core-image-sato-qemux86/201410131709/libgcc-initial-4.9.1-r0/do_extra_symlinks

EndTimeIO: 376316
StartTimeIO: 376312

So 0 != 4

Attached the db and the stat file.

Cheers,
Ionut C
Comment 1 Ionut Chisanovici 2014-10-14 09:03:47 UTC
Created attachment 2187 [details]
toaster.sqlite
Comment 2 Belen Barros Pena 2014-10-20 15:48:17 UTC
I am seeing this as well with a clean core-image-base on 
master:	924da20c6f2f183fa7e930e4c3c6b14c090acbe5

We should have a look at this to see what's going on.
Comment 3 Alexandru Damian 2014-11-03 15:20:40 UTC
Can't repro with the latest toaster.bbclass (patch at level "toaster.bbclass: read elapsed time from the stats file").

Please re-test after I put it on no-repro (I'll set once the patches are merged).
Comment 4 Ionut Chisanovici 2014-11-10 09:12:27 UTC
Hi Alex,

This issue appears in 1.6 as well.
Also, this is the disk_io issue and not the elapsed_time issue.

Cheers,
Ionut C
Comment 5 Alexandru Damian 2015-01-21 12:04:19 UTC
Please retest for 1.8 M3.
Comment 6 Alexandru Damian 2015-01-30 16:27:59 UTC
Works for me.
Comment 7 Ionut Chisanovici 2015-02-03 15:33:44 UTC
This issue also appears in 1.8 M2 RC2, commit: 22c38e2d8a058619e3d05534b5e830b750aa4241
Comment 8 Ionut Chisanovici 2015-02-26 14:00:11 UTC
Also, appears in 1.8_M3.RC1, commit: 40d80f9e019118ac7bcd268f21e350bc154a4a63.

Cheers,
Ionut C
Comment 9 Alexandru Damian 2015-03-19 18:48:16 UTC
I cannot reproduce this bug on 0308f3101b93035ebf564a33acb00adaac108033

Please attach all toaster*.log files in the build/ directory if you can reproduce it.
Comment 10 Alexandru Georgescu 2015-03-23 09:30:48 UTC
Mihai, Can you verify?
Comment 11 Mihail Stanciu 2015-03-23 12:05:23 UTC
Unable to reproduce on master: bbc139617637392a9b0fb2e9332a42ad64a4a556.

Will this be WNF for 1.6 and 1.7?
Comment 12 Belen Barros Pena 2015-04-10 14:38:55 UTC
Replaced with 7594, which tracks what's causing this problem.

*** This bug has been marked as a duplicate of bug 7594 ***
Comment 13 Alexandru Roman 2015-12-03 14:10:32 UTC
Verified Duplicate.