Nomad: cannot get status for jobs with a dash in job ID

[VascoVisser]

Nomad version

Nomad v0.6.1
Nomad v0.6.2

Operating system and Environment details

Linux 4.4.0-62-generic #83-Ubuntu SMP x86_64 GNU/Linux

Issue

Starting from nomad 0.6.1 I am not able to get the status for a job via the cli if that job has a dash in its name.

Reproduction steps

• download nomad 0.6.1 or 0.6.2
• start nomad: nomad agent -dev without any additional config
• run nomad status foo-bar

output:

Error querying search with id: "Unexpected response code: 500 (alloc lookup failed: index error: Input (without hyphens) must be even length)"

expected output (the same as when querying for a job ID without a dash):

Unable to resolve ID: "foo-bar"

Nomad Server logs (if appropriate)

    2017/08/31 00:15:40.243520 [ERR] http: Request /v1/search, error: alloc lookup failed: index error: Input (without hyphens) must be even length
    2017/08/31 00:15:40.244141 [DEBUG] http: Request /v1/search (818.813µs)

[dadgar]

Hey this has been fixed and will be out in 0.6.3. In the meantime please use nomad job status

[fho]

@dadgar the problem still happens with nomad 0.6.3.

[root@sb-fho-643 (screen) ~] nomad -v                                                                                                                                        
Nomad v0.6.3                                                                                                                                                                 
[root@sb-fho-643 (screen) ~] nomad status arregistry-service                                                                                                                 
Error querying search with id: "Unexpected response code: 500 (alloc lookup failed: index error: Input (without hyphens) must be even length)"                               
[root@sb-fho-643 (screen) ~] nomad status user-service  
Error querying search with id: "Unexpected response code: 500 (alloc lookup failed: index error: Input (without hyphens) must be even length)"                               

Can you reopen the ticket?

[dadgar]

@fho Sorry about that. Added tests to cover all possible values and have fixed this.

[metroidprototype]

I'm still seeing this in both 0.6.3 and 0.7.0-beta:

[rampal@latveria bin]$ nomad -v
Nomad v0.6.3
[rampal@latveria bin]$ nomad status splunk-forwarder
Error querying search with id: "Unexpected response code: 404 (404 page not found\n)"
[rampal@latveria bin]$ nomad -v
Nomad v0.7.0-beta1
[rampal@latveria bin]$ nomad status splunk-forwarder
Error querying search with id: "Unexpected response code: 404 (404 page not found\n)"
[rampal@latveria bin]$

Rolling my binary back to 0.5.6 works fine.

[dadgar]

@metroidprototype I believe you may have just upgraded the CLI binary and not the servers/clients as well. I just tried with the job name you gave on 0.7.0-beta1 and it worked correctly.