On Thu, 16 Nov 2017, Pavan Rallabhandi wrote: > Thanks for the reply Sage. On Jewel 10.2.10, I’ve just tested to find that if we mark the OSD manually out using `ceph osd out <id>`, the weights are getting reset once the OSD is back in the cluster. > > Instead, if I bring down the OSD and then let it get marked out by the default grace period of 300 seconds of it being down, it seems to work (meaning, the weights are persisted once the OSD is back in the cluster). > > Is that expected? Yeah. There was a change shortly after jewel to make the normal osd out preserve the weight the same way we do for auto-out osds. It's 73a2c1e. Could probably be backported, but it hasn't been. sage > > Thanks, > -Pavan. > > On 16/11/17, 6:39 PM, "Sage Weil" <sage@xxxxxxxxxxxx> wrote: > > On Thu, 16 Nov 2017, Pavan Rallabhandi wrote: > > Had to revive this old thread, had couple of questions. > > > > Since `ceph osd reweight-by-utilization` is changing the weights of the > > OSDs but not the CRUSH weights, is it still not a problem (as the OSD > > weights would be reset to 1) if those reweighted OSDs go OUT of the > > cluster and later get marked IN? > > > > I thought since OSD weights are not persistent across OUT/IN cycles, it > > is not a lasting solution to use `ceph osd reweight` or `ceph osd > > reweight-by-utilization`. > > This was fixed a while go, and a superficial check of the jewel code > indicates that the in/out values are persistent now. Have you observed > them getting reset with jewel? > > > We are having balancing issues on one of our Jewel clusters and I wanted > > to understand the pros of using `ceph osd reweight-by-utilization` over > > `ceph osd crush reweight`. > > Both will get the job done, but I would stick with reweight-by-utilization > as it keeps the real CRUSH weight matched to the device size. Once you > move to luminous, it will be an easier transition to the automatic > balancer module (which handles all of this for you). > > sage > > > > > > Thanks, > > -Pavan. > > > > From: Ceph-large <ceph-large-bounces@xxxxxxxxxxxxxx> on behalf of Dan Van Der Ster <daniel.vanderster@xxxxxxx> > > Date: Tuesday, 25 April 2017 at 11:36 PM > > To: Anthony D'Atri <aad@xxxxxxxxxxxxxx> > > Cc: "ceph-large@xxxxxxxxxxxxxx" <ceph-large@xxxxxxxxxxxxxx> > > Subject: EXT: Re: Rebalancing > > > > We run this continuously -- in a cron every 2 hours -- on all of our clusters: https://github.com/cernceph/ceph-scripts/blob/master/tools/crush-reweight-by-utilization.py > > It's a misnomer, yes -- because my original plan was indeed to modify CRUSH weights but for some reason which I do not recall, I switch it to modify the reweights. It should be super easy to change the crush weight instead. > > We run it with params to change weights of only 4 OSDs by 0.01 at a time. This ever so gradually flattens the PG distribution, and is totally transparent latency-wise. > > BTW, it supports reweighting only below certain CRUSH buckets, which is essential if you have a non-uniform OSD tree. > > > > For adding in new hardware, we use this script: https://github.com/cernceph/ceph-scripts/blob/master/tools/ceph-gentle-reweight > > New OSDs start with crush weight 0, then we gradually increase the weights 0.01 at a time, all the while watching the number of backfills and cluster latency. > > The same script is used to gradually drain OSDs down to CRUSH weight 0. > > We've used that second script to completely replace several petabytes of hardware. > > > > Cheers, Dan > > > > > > On 25 Apr 2017, at 08:22, Anthony D'Atri <aad@xxxxxxxxxxxxxx<mailto:aad@xxxxxxxxxxxxxx>> wrote: > > > > I read this thread with interest because I’ve been squeezing the OSD distirbution on several clusters mysel while expansion gear is in the pipline, ending up with an ugly mix of both types of reweight as well as temporarily raising the full and backfill full ratios. > > > > I’d been contemplating tweaking Dan@CERN’s reweighting script to use CRUSH reweighting instead, and to squeeze from both ends, though I fear it might not be as simple as it sounds prima fascia. > > > > > > Aaron wrote: > > > > > > Should I be expecting it to decide to increase some underutilized osds? > > > > > > The osd reweight mechanism only accomodates an override weight between 0 and 1, thus it can decrease but not increase a given OSD’s fullness. To directly fill up underfull OSD’s it would seem to to need an override weight > 1, which isn’t possible. > > > > I haven’t personally experienced it (yet), but from what I read, if override reweighted OSD’s get marked out and back in again, their override will revert to 1. In a case where a cluster is running close to the full ratio, this would *seem* as though a network glitch etc. might result in some OSD’s filling up and hitting the full threshold, which would be bad. > > > > Using CRUSH reweight instead would seem to address both of these shortcomings, though it does perturb the arbitrary but useful way that initial CRUSH weights by default reflect the capacity of each OSD. Various references also indicate that the override reweight does not change the weight of buckets above the OSD, but that CRUSH reweight does. I haven’t found any discussion of the ramifications of this, but my inital stab at it would be that when one does the 0-1 override reweight, the “extra’ data is redistributed to OSD’s on the same node. CRUSH reweighting would then seem to pull / push the wad of data being adjusted from / to *other* OSD nodes. Or it could be that I’m out of my Vulcan mind. > > > > Thus adjusting the weight of a given OSD affects the fullness of other OSD’s, in ways that would seem to differ depending on which method is used. As I think you implied in one of your messages, sometimes this can result in the fullness of one or more OSD’s climbing relatively sharply, even to a point distinctly above where the previous most-full OSDs were. > > > > I lurked in the recent developer’s meeting where strategies for A Better Way in Luminous were discussed. While the plans are exciting and hold promise for uniform and thus greater safe utilization of a cluster’s raw space, I suspect though that between dev/test time and the attrition needed to update running clients, those of us running existing RBD clusters won’t be able to take advantage of them for some time. > > > > — Anthony > > > > > > _______________________________________________ > > Ceph-large mailing list > > Ceph-large@xxxxxxxxxxxxxx<mailto:Ceph-large@xxxxxxxxxxxxxx> > > http://lists.ceph.com/listinfo.cgi/ceph-large-ceph.com > > > > > > > >
_______________________________________________ Ceph-large mailing list Ceph-large@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-large-ceph.com
