juju machine agent runtime panic
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
| juju |
High
|
John A Meinel | ||
| 2.8 |
High
|
John A Meinel |
Bug Description
We observed a juju machine agent panic:
2020-11-05 19:14:49 WARNING juju.apiserver.
2020-11-05 19:14:51 ERROR juju.worker.
2020-11-05 19:14:51 WARNING juju.apiserver.
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0xdbd38d]
goroutine 8771166 [running]:
github.
created by github.
2020-11-05 19:14:53 INFO juju.cmd supercommand.go:54 running jujud [2.8.3 0 ab69570b38fbc74
2020-11-05 19:14:53 DEBUG juju.cmd supercommand.go:55 args: []string{
Let me know if there is any additional information I can provide.
John A Meinel (jameinel) wrote : | #1 |
Changed in juju: | |
status: | New → Triaged |
importance: | Undecided → High |
John A Meinel (jameinel) wrote : | #2 |
Changed in juju: | |
milestone: | none → 2.9-rc3 |
assignee: | nobody → John A Meinel (jameinel) |
status: | Triaged → In Progress |
Colin Misare (cmisare) wrote : | #3 |
We saw a second instance of this issue on a different machine in the same environment. Is it possible something in this environment is exacerbating the transaction behavior causing these nil pointers to appear more frequently?
2020-11-06 18:51:01 ERROR juju.worker.
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0xdbd38d]
goroutine 27078232 [running]:
github.
created by github.
2020-11-06 18:51:13 INFO juju.cmd supercommand.go:54 running jujud [2.8.3 0 ab69570b38fbc74
2020-11-06 18:51:13 DEBUG juju.cmd supercommand.go:55 args: []string{
2020-11-06 18:51:13 DEBUG juju.utils gomaxprocs.go:24 setting GOMAXPROCS to 4
2020-11-06 18:51:13 DEBUG juju.agent agent.go:583 read agent config, format "2.0"
2020-11-06 18:51:13 INFO juju.cmd.jujud agent.go:138 setting logging config to "<root>
It happens if we get an error will issuing a removeAll command. It is
possible that the list of transactions has gotten large enough it is
causing problems for pruning to clean it up, or there is some other sort of
interaction that is causing more problems.
On Fri, Nov 6, 2020 at 3:45 PM Colin Misare <email address hidden>
wrote:
> We saw a second instance of this issue on a different machine in the
> same environment. Is it possible something in this environment is
> exacerbating the transaction behavior causing these nil pointers to
> appear more frequently?
>
>
> 2020-11-06 18:51:01 ERROR juju.worker.
> "firewaller" manifold worker returned unexpected error: machine 5 not
> provisioned
> panic: runtime error: invalid memory address or nil pointer dereference
> [signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0xdbd38d]
>
> goroutine 27078232 [running]:
> github.
> 0xc011a7a000, 0x3e8, 0x3e8, 0xc010992780, 0xc01ac30fc0, 0xc0114cd0e0,
> 0xc00c522200)
> /workspace/
> github.
> +0x1ed
> created by github.
> /workspace/
> github.
> +0x161
> 2020-11-06 18:51:13 INFO juju.cmd supercommand.go:54 running jujud [2.8.3
> 0 ab69570b38fbc74
> 2020-11-06 18:51:13 DEBUG juju.cmd supercommand.go:55 args:
> []string{
> "/var/lib/juju", "--machine-id", "2", "--debug"}
> 2020-11-06 18:51:13 DEBUG juju.utils gomaxprocs.go:24 setting GOMAXPROCS
> to 4
> 2020-11-06 18:51:13 DEBUG juju.agent agent.go:583 read agent config,
> format "2.0"
> 2020-11-06 18:51:13 INFO juju.cmd.jujud agent.go:138 setting logging
> config to "<root>
>
> --
> You received this bug notification because you are a bug assignee.
> Matching subscriptions: juju bugs
> https:/
>
> Title:
> juju machine agent runtime panic
>
> To manage notifications about this bug go to:
> https:/
>
Changed in juju: | |
status: | In Progress → Fix Committed |
https:/ /github. com/juju/ txn/pull/ 57 is a possible fix in the lower library.
Essentially an error at exactly the right time when trying to remove old transactions will bubble up and we end up accessing a nil pointer.