Difference between revisions of "GGUS:Ticket monitoring"

From EGIWiki
Jump to: navigation, search
 
(18 intermediate revisions by 4 users not shown)
Line 1: Line 1:
{{GGUS-header|GGUS ticket monitoring}}
+
{{GGUS-header|GGUS ticket monitoring}}  
<span style="color:red">This page is in draft version!</span> <br><br>
 
= Ticket Monitoring Dashboard =
 
For easier triage of critical tickets a [https://ggus.eu/pages/monitor.php dashboard] was implemented.
 
This dashboard is linked from the "GGUS tools/reports" section with label "Monitor" on [https://ggus.eu/pages/home.php GGUS Home].
 
  
== Operations Tickets ==
+
== Ticket Monitoring Dashboard  ==
Operations tickets are submitted via the operations portal. They have ticket type '''OPS'''. Monitoring of '''OPS''' tickets is done by the COD team.
 
  
== Middleware Tickets ==
+
For easier triage of critical tickets a [https://ggus.eu/?mode=ticket_monitor dashboard] was implemented. This dashboard is linked from the "GGUS tools/reports" section with label "Monitor" on [https://ggus.eu/?mode=index GGUS Home].<br> The dashboard is only visible for the GGUS monitoring group.
* Please see [[EGI_DMSU_Ticket_Followup]] before taking any action.
 
* Useful information available on [https://rt.egi.eu/rt/Dashboards/4248/Software%20Provisioning%20UMD-2 UMD Dashboard]
 
  
== Other Tickets ==
+
== Operations Tickets ==
The monitoring process is documented in [[File:Other_Tickets.pdf]]
 
  
<!--
+
Operations tickets are submitted via the operations portal. They are of ticket type '''OPS'''. <br> Monitoring of '''OPS''' tickets is done by the NGIs ROD teams.  
GGUS ticket monitoring and follow up after July 1st 2012.
 
  
== Rules ==
+
== Middleware Tickets  ==
* Rules will be generated from the ideas below.
 
  
== Ideas ==
+
The monitoring of middleware tickets follows the processes described in this [https://wiki.egi.eu/wiki/FAQ_GGUS-Waiting-For-PT-Process FAQ].  
* Implement a dashboard for easier triage of (critical) tickets
 
* Submit tickets against [[GGUS:Operations FAQ|"Operations" SU]] listing untouched tickets of a dedicated SU
 
* Technology Tickets: in case the ETA (Estimated Time of Availability) is set the ticket monitoring team will not act on the ticket before the ETA date.
 
  
=== Closing old tickets ===
+
*Relevant information is also available here:
''Old tickets'' are tickets opened before July 1st 2012. (''XX'' and ''YY'' needs to be defined below!)
+
**[[EGI DMSU Ticket Followup]]
 +
**[https://rt.egi.eu/rt/Dashboards/6388/Software%20Provisioning%20UMD-4 UMD 4], [https://rt.egi.eu/rt/Dashboards/10379/Software%20Provisioning%20CMD-OS-1 CMD-OS 1] Dashboards
 +
**[[Middleware#Technology_Providers]]
 +
*The assignment of ETA is checked by DMSU. This process is still valid.
 +
*When the ETA date/time arrives, the ticket monitoring checks whether the fix was delivered and is part of any UMD release.
  
* A ticket opened before July 1st 2012 and being untouched for ''XX'' days.
+
In general the monitoring process given in [[Image:Ticket Monitoring.pdf]] is also valid for middleware tickets.  
*# will be put in status "waiting for reply" to see, if the ticket opener still needs help.
 
*# when the ticket opener replies in ''YY'' days:  
 
*##TPM will contact the SU about the ticket.
 
*## if not, the ticket will be closed. The user can verify or reopen the ticket afterwards.
 
* Check whether ticket is related to another ticket in any bug tracker
 
** Check ticket in bug tracker if possible -> sometimes access impossible
 
** Compare "planned release date" in bug tracker with UMD repository
 
  
=== Monitoring new tickets ===
+
*If there are doubts about tickets, please submit a ticket to '''Operations''' SU referencing the tickets in question. '''Operations''' SU will then escalate this to TCB
''New tickets'' are tickets opened before July 1st 2012.
 
  
==== Middleware tickets ====
+
== Other Tickets ==
*# Please see [[EGI_DMSU_Ticket_Followup]] before taking any action.
 
*# Useful information available on [https://rt.egi.eu/rt/Dashboards/4248/Software%20Provisioning%20UMD-2 UMD Dashboard]
 
*# Escalation time
 
*#* 4 weeks untouched
 
*# Escalation steps
 
*#* first add comment to ticket
 
*#* create new ticket with management information referencing ticket in question and assign it to
 
*#** [[GGUS:EMI_FAQ|EMI SU]] in case of EMI middleware problems
 
*#** ?? in case of IGE middleware problems
 
*#** ?? in case of SAGA middleware problems
 
==== Operations Tickets ====
 
*# Escalation time
 
*#* 2 weeks untouched
 
*# Escalation steps
 
*#* first add comment to ticket
 
*#* create new ticket with management information referencing ticket in question and assign it to
 
*#** NGI in case of site problems
 
*#** [[GGUS:Operations FAQ|EGI Operations]] in case of other problems
 
*#* in case of NGIs not responding assign the relevant ticket to COD. COD will contact the operations manager directly. If the COD escalation step fails, COD moves the ticket to the [[GGUS:Operations FAQ|EGI Operations]] SU.
 
* Tickets for other Infrastructures
 
* Tickets waiting for user reply will be closed after 5 working days.
 
** example: https://ggus.eu/ws/ticket_info.php?ticket=84557
 
  
=== Handling of ticket escalation level 2 ===
+
The monitoring process for non-operations and non-middleware tickets is documented in [[Image:Ticket Monitoring.pdf]] in detail. A short summary of required actions is as follows:
* Tickets can be escalated by the submitter. (already implemented)
 
  
=== Double Tickets ===
+
*add comments
* Users can duplicate a '''Savannah ticket''' into one or more GGUS tickets by answering savannah mails
+
*tag ticket for adding to monitoring dashboard
** The GGUS ticket can be closed immediately, if savannah status is: "Use GGUS: No" and the answer of the user is in Savannah too.
+
*set follow-up date
** example https://ggus.eu/ws/ticket_info.php?ticket=85287
+
*set appropriate ''remind on'' date
=== Slave Tickets ===
+
*assign ticket to site operations managers if applicable
* if a TPM ticket becomes a slave of another ticket before it is assigned to the correct SU, the ticket remains as "TPM on hold". In this case
+
*involve NGI operations managers if necessary
*# TPM will remove the master slave relation
+
*submit a ticket against "Operations" SU referencing untouched tickets
*# TPM assigns the slave tickets to the correct SU
+
 
*# TPM restores the master slave relation
+
== References  ==
* Example https://ggus.eu/ws/ticket_info.php?ticket=85380
+
 
-->
+
<references />

Latest revision as of 13:34, 22 October 2018

GGUS-logo.jpg


GGUS wiki / GGUS Documentation


GGUS ticket monitoring



Ticket Monitoring Dashboard

For easier triage of critical tickets a dashboard was implemented. This dashboard is linked from the "GGUS tools/reports" section with label "Monitor" on GGUS Home.
The dashboard is only visible for the GGUS monitoring group.

Operations Tickets

Operations tickets are submitted via the operations portal. They are of ticket type OPS.
Monitoring of OPS tickets is done by the NGIs ROD teams.

Middleware Tickets

The monitoring of middleware tickets follows the processes described in this FAQ.

In general the monitoring process given in File:Ticket Monitoring.pdf is also valid for middleware tickets.

  • If there are doubts about tickets, please submit a ticket to Operations SU referencing the tickets in question. Operations SU will then escalate this to TCB

Other Tickets

The monitoring process for non-operations and non-middleware tickets is documented in File:Ticket Monitoring.pdf in detail. A short summary of required actions is as follows:

  • add comments
  • tag ticket for adding to monitoring dashboard
  • set follow-up date
  • set appropriate remind on date
  • assign ticket to site operations managers if applicable
  • involve NGI operations managers if necessary
  • submit a ticket against "Operations" SU referencing untouched tickets

References