Re: Bug in crush algorithm? 1 PG with the same OSD twice.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



BTW, I vaguely recalled seeing this before. Yup, found it:
https://tracker.ceph.com/issues/55169

On Tue, Aug 30, 2022 at 11:46 AM Dan van der Ster <dvanders@xxxxxxxxx> wrote:
>
> > 2. osd.7 is destroyed but still "up" in the osdmap.
>
> Oops, you can ignore this point -- this was an observation I had while
> playing with the osdmap -- your osdmap.bin has osd.7 down correctly.
>
> In case you're curious, here was what confused me:
>
> # osdmaptool osdmap.bin2  --mark-up-in --mark-out 7 --dump plain
> osd.7 up   out weight 0 up_from 3846 up_thru 3853 down_at 3855
> last_clean_interval [0,0)
> [v2:10.41.24.15:6810/1915819,v1:10.41.24.15:6811/1915819]
> [v2:192.168.0.15:6808/1915819,v1:192.168.0.15:6809/1915819]
> destroyed,exists,up
>
> Just ignore this ...
>
>
>
> -- dan
>
> On Tue, Aug 30, 2022 at 11:41 AM Dan van der Ster <dvanders@xxxxxxxxx> wrote:
> >
> > Hi Frank,
> >
> > I suspect this is a combination of issues.
> > 1. You have "choose" instead of "chooseleaf" in rule 1.
> > 2. osd.7 is destroyed but still "up" in the osdmap.
> > 3. The _tries settings in rule 1 are not helping.
> >
> > Here are my tests:
> >
> > # osdmaptool --test-map-pg 4.1c osdmap.bin
> > osdmaptool: osdmap file 'osdmap.bin'
> >  parsed '4.1c' -> 4.1c
> > 4.1c raw ([6,1,4,5,3,2147483647], p6) up ([6,1,4,5,3,2147483647], p6)
> > acting ([6,1,4,5,3,1], p6)
> >
> > ^^ This is what you observe now.
> >
> > # diff -u crush.txt crush.txt2
> > --- crush.txt 2022-08-30 11:27:41.941836374 +0200
> > +++ crush.txt2 2022-08-30 11:31:29.631491424 +0200
> > @@ -93,7 +93,7 @@
> >   step set_chooseleaf_tries 50
> >   step set_choose_tries 200
> >   step take default
> > - step choose indep 0 type osd
> > + step chooseleaf indep 0 type osd
> >   step emit
> >  }
> > # crushtool -c crush.txt2 -o crush.map2
> > # cp osdmap.bin osdmap.bin2
> > # osdmaptool --import-crush crush.map2 osdmap.bin2
> > osdmaptool: osdmap file 'osdmap.bin2'
> > osdmaptool: imported 1166 byte crush map from crush.map2
> > osdmaptool: writing epoch 4990 to osdmap.bin2
> > # osdmaptool --test-map-pg 4.1c osdmap.bin2
> > osdmaptool: osdmap file 'osdmap.bin2'
> >  parsed '4.1c' -> 4.1c
> > 4.1c raw ([6,1,4,5,3,7], p6) up ([6,1,4,5,3,2147483647], p6) acting
> > ([6,1,4,5,3,1], p6)
> >
> > ^^ The mapping is now "correct" in that it doesn't duplicate the
> > mapping to osd.1. However it tries to use osd.7 which is destroyed but
> > up.
> >
> > You might be able to fix that by fully marking osd.7 out.
> > I can also get a good mapping by removing the *_tries settings from rule 1:
> >
> > # diff -u crush.txt crush.txt2
> > --- crush.txt 2022-08-30 11:27:41.941836374 +0200
> > +++ crush.txt2 2022-08-30 11:38:14.068102835 +0200
> > @@ -90,10 +90,8 @@
> >   type erasure
> >   min_size 3
> >   max_size 6
> > - step set_chooseleaf_tries 50
> > - step set_choose_tries 200
> >   step take default
> > - step choose indep 0 type osd
> > + step chooseleaf indep 0 type osd
> >   step emit
> >  }
> > ...
> > # osdmaptool --test-map-pg 4.1c osdmap.bin2
> > osdmaptool: osdmap file 'osdmap.bin2'
> >  parsed '4.1c' -> 4.1c
> > 4.1c raw ([6,1,4,5,3,2], p6) up ([6,1,4,5,3,2], p6) acting ([6,1,4,5,3,1], p6)
> >
> > Note that I didn't need to adjust the reweights:
> >
> > # osdmaptool osdmap.bin2 --tree
> > osdmaptool: osdmap file 'osdmap.bin2'
> > ID CLASS WEIGHT  TYPE NAME         STATUS    REWEIGHT PRI-AFF
> > -1       2.44798 root default
> > -7       0.81599     host tceph-01
> >  0   hdd 0.27199         osd.0            up  0.87999 1.00000
> >  3   hdd 0.27199         osd.3            up  0.98000 1.00000
> >  6   hdd 0.27199         osd.6            up  0.92999 1.00000
> > -3       0.81599     host tceph-02
> >  2   hdd 0.27199         osd.2            up  0.95999 1.00000
> >  4   hdd 0.27199         osd.4            up  0.89999 1.00000
> >  8   hdd 0.27199         osd.8            up  0.89999 1.00000
> > -5       0.81599     host tceph-03
> >  1   hdd 0.27199         osd.1            up  0.89999 1.00000
> >  5   hdd 0.27199         osd.5            up  1.00000 1.00000
> >  7   hdd 0.27199         osd.7     destroyed        0 1.00000
> >
> >
> > Does this work in real life?
> >
> > Cheers, Dan
> >
> >
> > On Mon, Aug 29, 2022 at 7:38 PM Frank Schilder <frans@xxxxxx> wrote:
> > >
> > > Hi Dan,
> > >
> > > please find attached (only 7K, so I hope it goes through). md5sum=1504652f1b95802a9f2fe3725bf1336e
> > >
> > > I was playing a bit around with the crush map and found out the following:
> > >
> > > 1) Setting all re-weights to 1 does produce valid mappings. However, it will lead to large imbalances and is impractical in operations.
> > >
> > > 2) Doing something as simple/stupid as the following also results in valid mappings without having to change the weights:
> > >
> > > rule fs-data {
> > >         id 1
> > >         type erasure
> > >         min_size 3
> > >         max_size 6
> > >         step set_chooseleaf_tries 50
> > >         step set_choose_tries 200
> > >         step take default
> > >         step chooseleaf indep 3 type host
> > >         step emit
> > >         step take default
> > >         step chooseleaf indep -3 type host
> > >         step emit
> > > }
> > >
> > > rule fs-data {
> > >         id 1
> > >         type erasure
> > >         min_size 3
> > >         max_size 6
> > >         step set_chooseleaf_tries 50
> > >         step set_choose_tries 200
> > >         step take default
> > >         step choose indep 3 type osd
> > >         step emit
> > >         step take default
> > >         step choose indep -3 type osd
> > >         step emit
> > > }
> > >
> > > Of course, now the current weights are probably unsuitable as everything moves around. Its probably also a lot more total tries to get rid of mappings with duplicate OSDs.
> > >
> > > I probably have to read the code to understand how drawing straws from 8 different buckets with non-zero probabilities can lead to an infinite sequence of failed attempts of getting 6 different ones. There seems to be a hard-coded tunable that turns seemingly infinite into finite somehow.
> > >
> > > The first modified rule will probably lead to better distribution of load, but bad distribution of data if a disk goes down (considering the tiny host- and disk numbers). The second rule seems to be almost as good or bad as the default one (step choose indep 0 type osd), except that it does produce valid mappings where the default rule fails.
> > >
> > > I will wait with changing the rule in the hope that you find a more elegant solution to this riddle.
> > >
> > > Best regards,
> > > =================
> > > Frank Schilder
> > > AIT Risø Campus
> > > Bygning 109, rum S14
> > >
> > > ________________________________________
> > > From: Dan van der Ster <dvanders@xxxxxxxxx>
> > > Sent: 29 August 2022 19:13
> > > To: Frank Schilder
> > > Subject: Re:  Bug in crush algorithm? 1 PG with the same OSD twice.
> > >
> > > Hi Frank,
> > >
> > > Could you share the osdmap so I can try to solve this riddle?
> > >
> > > Cheers , Dan
> > >
> > >
> > > On Mon, Aug 29, 2022, 17:26 Frank Schilder <frans@xxxxxx<mailto:frans@xxxxxx>> wrote:
> > > Hi Dan,
> > >
> > > thanks for your answer. I'm not really convinced that we hit a corner case here and even if its one, it seems quite relevant for production clusters. The usual way to get a valid mapping is to increase the number of tries. I increased the following max trial numbers, which I would expect to produce a mapping for all PGs:
> > >
> > > # diff map-now.txt map-new.txt
> > > 4c4
> > > < tunable choose_total_tries 50
> > > ---
> > > > tunable choose_total_tries 250
> > > 93,94c93,94
> > > <       step set_chooseleaf_tries 5
> > > <       step set_choose_tries 100
> > > ---
> > > >       step set_chooseleaf_tries 50
> > > >       step set_choose_tries 200
> > >
> > > When I test the map with crushtool it does not report bad mappings. Am I looking at the wrong tunables to increase? It should be possible to get valid mappings without having to modify the re-weights.
> > >
> > > Thanks again for your help!
> > > =================
> > > Frank Schilder
> > > AIT Risø Campus
> > > Bygning 109, rum S14
> > >
> > > ________________________________________
> > > From: Dan van der Ster <dvanders@xxxxxxxxx<mailto:dvanders@xxxxxxxxx>>
> > > Sent: 29 August 2022 16:52:52
> > > To: Frank Schilder
> > > Cc: ceph-users@xxxxxxx<mailto:ceph-users@xxxxxxx>
> > > Subject: Re:  Bug in crush algorithm? 1 PG with the same OSD twice.
> > >
> > > Hi Frank,
> > >
> > > CRUSH can only find 5 OSDs, given your current tree, rule, and
> > > reweights. This is why there is a NONE in the UP set for shard 6.
> > > But in ACTING we see that it is refusing to remove shard 6 from osd.1
> > > -- that is the only copy of that shard, so in this case it's helping
> > > you rather than deleting the shard altogether.
> > > ACTING == what the OSDs are serving now.
> > > UP == where CRUSH wants to place the shards.
> > >
> > > I suspect that this is a case of CRUSH tunables + your reweights
> > > putting CRUSH in a corner case of not finding 6 OSDs for that
> > > particular PG.
> > > If you set the reweights all back to 1, it probably finds 6 OSDs?
> > >
> > > Cheers, Dan
> > >
> > >
> > > On Mon, Aug 29, 2022 at 4:44 PM Frank Schilder <frans@xxxxxx<mailto:frans@xxxxxx>> wrote:
> > > >
> > > > Hi all,
> > > >
> > > > I'm investigating a problem with a degenerated PG on an octopus 15.2.16 test cluster. It has 3Hosts x 3OSDs and a 4+2 EC pool with failure domain OSD. After simulating a disk fail by removing an OSD and letting the cluster recover (all under load), I end up with a PG with the same OSD allocated twice:
> > > >
> > > > PG 4.1c, UP: [6,1,4,5,3,NONE] ACTING: [6,1,4,5,3,1]
> > > >
> > > > OSD 1 is allocated twice. How is this even possible?
> > > >
> > > > Here the OSD tree:
> > > >
> > > > ID  CLASS  WEIGHT   TYPE NAME          STATUS     REWEIGHT  PRI-AFF
> > > > -1         2.44798  root default
> > > > -7         0.81599      host tceph-01
> > > >  0    hdd  0.27199          osd.0             up   0.87999  1.00000
> > > >  3    hdd  0.27199          osd.3             up   0.98000  1.00000
> > > >  6    hdd  0.27199          osd.6             up   0.92999  1.00000
> > > > -3         0.81599      host tceph-02
> > > >  2    hdd  0.27199          osd.2             up   0.95999  1.00000
> > > >  4    hdd  0.27199          osd.4             up   0.89999  1.00000
> > > >  8    hdd  0.27199          osd.8             up   0.89999  1.00000
> > > > -5         0.81599      host tceph-03
> > > >  1    hdd  0.27199          osd.1             up   0.89999  1.00000
> > > >  5    hdd  0.27199          osd.5             up   1.00000  1.00000
> > > >  7    hdd  0.27199          osd.7      destroyed         0  1.00000
> > > >
> > > > I tried already to change some tunables thinking about https://docs.ceph.com/en/octopus/rados/troubleshooting/troubleshooting-pg/#crush-gives-up-too-soon, but giving up too soon is obviously not the problem. It is accepting a wrong mapping.
> > > >
> > > > Is there a way out of this? Clearly this is calling for trouble if not data loss and should not happen at all.
> > > >
> > > > Best regards,
> > > > =================
> > > > Frank Schilder
> > > > AIT Risø Campus
> > > > Bygning 109, rum S14
> > > > _______________________________________________
> > > > ceph-users mailing list -- ceph-users@xxxxxxx<mailto:ceph-users@xxxxxxx>
> > > > To unsubscribe send an email to ceph-users-leave@xxxxxxx<mailto:ceph-users-leave@xxxxxxx>
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux