Reports until 12:25, Monday 11 November 2019
H1 CDS
david.barker@LIGO.ORG - posted 12:25, Monday 11 November 2019 - last comment - 15:14, Monday 11 November 2019(53159)
Deleted old ext_alert services which were competing with lv_alert

TJ, Jonathan, Dave:

On the machine ext-alert (note dash, not underscore) we were running the legacy ext_alert processes alongside the new lv_alert. The legacy code was competing with the new causing GraceDB query errors.

I deleted the two services:

leaving just the lv_alert.service running. The deletion process I used is:

  1. systemctl stop service_name
  2. systemctl disable service_name
  3. mv /etc/systemd/system/service_name /etc/systemd/archive/service_name_11nov2019
  4. systemctl daemon-reload
  5. systemctl reset-failed
Comments related to this report
david.barker@LIGO.ORG - 15:14, Monday 11 November 2019 (53161)
Link

These services were subsequently restarted by puppet. Jonathan made the necessary manifest changes and these superfluous services are once again gone, this time hopefully for ever.