Crash while parsing /proc/stat

[DSpeichert]

panic: runtime error: slice bounds out of range

goroutine 35 [running]:
github.com/prometheus/node_exporter/collector.(*statCollector).Update(0xc20806e200, 0xc2080526c0, 0x0, 0x0)
        /usr/local/src/node_exporter/.deps/gopath/src/github.com/prometheus/node_exporter/collector/stat.go:101 +0x74a
main.Execute(0x8aef80, 0x4, 0x7f716c18a4d0, 0xc20806e200, 0xc2080526c0)
        /usr/local/src/node_exporter/node_exporter.go:89 +0x75
main.func·001(0x8aef80, 0x4, 0x7f716c18a4d0, 0xc20806e200)
        /usr/local/src/node_exporter/node_exporter.go:62 +0x5b
created by main.NodeCollector.Collect
        /usr/local/src/node_exporter/node_exporter.go:64 +0x1c7

Kernel 2.6.32-25-pve provides 8 instead of 9 values for each CPU.

[juliusv]

Thanks so much for sending in a fix! Any chance you could provide us with a sample output (in a gist or similar) of your /proc/stat? That way we can write a regression test (unless you feel like doing one).

[DSpeichert]

It's very short, here it is all:

cpu  99861709 51704 39800352 4643329336 2997662616 0 0 8316789
cpu0 38789986 3414 18464189 1123843679 1633841492 0 0 4026454
cpu1 24391091 8126 9189852 1167806616 406578316 0 0 1767912
cpu2 19355490 18228 6877332 1173997654 518819897 0 0 1365570
cpu3 17325141 21934 5268977 1177681384 438422909 0 0 1156851
intr 0
swap 0 0

ctxt 114984417539
btime 1411063515
processes 6868070
procs_running 1
procs_blocked 1

Important note - I just realized that the output is different INSIDE an OpenVZ guest (as above) than on the hypervisor. Hypervisor has all 9 values: https://gist.github.com/DSpeichert/6667c1998873cbb5e77b

[juliusv]

That's great to know - thanks! I'm guessing this will affect other paravirtualized guests as well (full VMs won't know that they're VMs...)

[juliusv]

(@grobie now working on moving this /proc/stat parser into https://github.com/prometheus/procfs and adding test there - the node_exporter will then simply use that library)