Project

General

Profile

Bug #3693

crushtool compile fails with unhelpful message, diagnosis quite difficult

Added by Dan Mick about 11 years ago. Updated almost 7 years ago.

Status:
Duplicate
Priority:
Low
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
Development
Tags:
Backport:
Regression:
No
Severity:
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(RADOS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

A user tried to create his own crushmap as follows:

# begin crush map

# devices
device 10 osd.10
device 11 osd.11
device 12 osd.12
device 13 osd.13
device 14 osd.14
device 15 osd.15
device 16 osd.16
device 17 osd.17
device 20 osd.20
device 21 osd.21
device 22 osd.22
device 23 osd.23
device 24 osd.24
device 25 osd.25
device 26 osd.26
device 27 osd.27

# types
type 0 osd
type 1 host
type 2 rack
type 3 net
type 4 room
type 5 datacenter
type 6 root

host keron {
    id -2
    alg straw
    hash 0    # rjenkins1
    item osd.10 weight 1.000
    item osd.11 weight 1.000
    item osd.12 weight 1.000
    item osd.13 weight 1.000
    item osd.14 weight 1.000
    item osd.15 weight 1.000
    item osd.16 weight 1.000
    item osd.17 weight 1.000
}
host brontes {
    id -3
    alg straw
    hash 0    # rjenkins1
    item osd.20 weight 0
    item osd.21 weight 0
    item osd.22 weight 0
    item osd.23 weight 0
    item osd.24 weight 0
    item osd.25 weight 0
    item osd.26 weight 0
    item osd.27 weight 0
}

rack 15B01 {
    id -40
    alg straw
    hash 0    # rjenkins1
    item brontes weight 0
}
rack 27A04 {
    id -43
    alg straw
    hash 0    # rjenkins1
    item keron weight 8
}

net 188-165-14 {
    id -71
    alg straw
    hash 0    # rjenkins1
    item rack 27A04 weight 8
}
net 188-165-15 {
    id -72
    alg straw
    hash 0    # rjenkins1
    item rack 15B01 weight 0
}

room S15 {
    id -90
    alg straw
    hash 0    # rjenkins1
    item net 188-165-15 weight 0
}
room S27 {
    id -91
    alg straw
    hash 0    # rjenkins1
    item net 188-165-14 weight 8
}

datacenter RBX1 {
    id -100
    alg straw
    hash 0    # rjenkins1
    item room S15 weight 0
}
datacenter RBX2 {
    id -101
    alg straw
    hash 0    # rjenkins1
    item room S27 weight 8
}

root default {
    id -1
    alg straw
    hash 0    # rjenkins1
    item datacenter RBX1 weight 0
    item datacenter RBX2 weight 8
}

# rules
rule data {
    ruleset 0
    type replicated
    min_size 1
    max_size 10
    step take default
    step choose firstn 0 type osd
    step emit
}
rule metadata {
    ruleset 1
    type replicated
    min_size 1
    max_size 10
    step take default
    step choose firstn 0 type osd
    step emit
}
rule rbd {
    ruleset 2
    type replicated
    min_size 1
    max_size 10
    step take default
    step choose firstn 0 type osd
    step emit
}

# end crush map

crushtool -c shows for this:

kioob.crushmap:70 error: parse error at ''

and various permutations on the input don't help narrow down the problem, which is, according
to Caleb:

the problem was that he specified a type along with the name when building a bucket, for example

root default {
        id -1
        alg straw
        hash 0  # rjenkins1
        item RBX1 weight 0
        item RBX2 weight 8
}

is ok, but

root default {
        id -1
        alg straw
        hash 0  # rjenkins1
        item datacenter RBX1 weight 0
        item datacenter RBX2 weight 8
}

will create an error


Related issues

Related to Ceph - Bug #3809: crush compiler errors are not helpful Won't Fix 01/15/2013

History

#1 Updated by Sage Weil about 11 years ago

  • Assignee set to caleb miles

#2 Updated by Anonymous almost 11 years ago

  • Project changed from Ceph to Calamari

#3 Updated by Anonymous almost 11 years ago

  • Assignee changed from caleb miles to Dan Mick

#4 Updated by Ian Colle almost 11 years ago

  • Project changed from Calamari to Ceph

#5 Updated by Dan Mick over 10 years ago

  • Subject changed from cephtool compile fails with unhelpful message, diagnosis quite difficult to crushtool compile fails with unhelpful message, diagnosis quite difficult
  • Category set to 10

#6 Updated by Dan Mick over 10 years ago

  • Assignee deleted (Dan Mick)

Removing myself as owner; there's no way I'm going to be able to untangle boost::spirit in anything less than months

#7 Updated by Sage Weil over 10 years ago

  • Priority changed from Normal to Low

#8 Updated by Sage Weil about 9 years ago

  • Status changed from New to Duplicate

#9 Updated by Greg Farnum almost 7 years ago

  • Project changed from Ceph to RADOS
  • Category deleted (10)

Also available in: Atom PDF