mpana/spearhead-issue-response
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

refactor to sphs specifics

Browse Source
This commit is contained in:
Marius Pana 2017-01-21 14:30:24 +02:00
parent 5b4f9a148f
commit e85a239ff8
1 changed files with 15 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
docs/after/post_mortem_process.md
View File

@ -1,24 +1,24 @@
For every major incident (SEV-2/1), we need to follow up with a post-mortem. A blame-free, detailed description, of exactly what went wrong in order to cause the incident, along with a list of steps to take in order to prevent a similar incident from occurring again in the future. The incident response process itself should also be included. For every major issue (SR/IN +major), we need to follow up with a post-mortem. A blame-free, detailed description, of exactly what went wrong in order to cause the incident, along with a list of steps to take in order to prevent a similar incident from occurring again in the future. The incident response process itself should also be included.
![Post-Mortem](../assets/img/headers/pagerduty_post_mortem.jpg) ![Post-Mortem](../assets/img/headers/pagerduty_post_mortem.jpg)
## Owner Designation ## Owner Designation
The first step is that a post-mortem owner will be designated. This is done by the IC either at the end of a major incident call, or very shortly after. You will be notified directly by the IC if you are the owner for the post-mortem. The owner is responsible for populating the post-mortem page, looking up logs, managing the followup investigation, and keeping all interested parties in the loop. Please use Slack for coordinating followup. A detailed list of the steps is available below, The first step is that a post-mortem owner will be designated. This is done by the TL either at the end of a major incident call, or very shortly after. You will be notified directly by the TL if you are the owner for the post-mortem. The owner is responsible for populating the post-mortem page, looking up logs, managing the followup investigation, and keeping all interested parties in the loop. Please use DoIT and Slack for coordinating followup. A detailed list of the steps is available below,
## Owner Responsibilities ## Owner Responsibilities
As owner of a post-mortem, you are responsible for the following, As owner of a post-mortem, you are responsible for the following,
* Scheduling the post-mortem meeting (on the shared calendar) and inviting the relevant people (this should be scheduled within 5 business days of the incident). * Scheduling the post-mortem meeting (on a shared calendar) and inviting the relevant people (this should be scheduled within 5 business days of the incident).
* Updating the page with all of the necessary content. * Updating the page with all of the necessary content.
* Investigating the incident, pulling in whomever you need from other teams to assist in the investigation. * Investigating the incident, pulling in whomever you need from other teams to assist in the investigation.
* Creating follow-up JIRA tickets (_You are only responsible for creating the tickets, not following them up to resolution_). * Creating follow-up DoIT cards (_You are only responsible for creating the cards, not following them up to resolution_).
* Running the post-mortem meeting (_these generally run themselves, but you should get people back on topic if the conversation starts to wander_). * Running the post-mortem meeting (_these generally run themselves, but you should get people back on topic if the conversation starts to wander_).
* In cases where we need a public blog post, creating & reviewing it with appropriate parties. * In cases where we need a public blog post, creating & reviewing it with appropriate parties.
## Post-Mortem Wiki Page ## Post-Mortem Wiki Page
Once you've been designated as the owner of a post-mortem, you should start updating the page with all the relevant information. Once you've been designated as the owner of a post-mortem, you should start updating the page with all the relevant information.
1. (If not already done by the IC) Create a new post-mortem page for the incident. 1. (If not already done by the TL) Create a new post-mortem page for the incident.
1. Schedule a post-mortem meeting for within 5 business days of the incident. You should schedule this before filling in the page, just so it's on the calendar. 1. Schedule a post-mortem meeting for within 5 business days of the incident. You should schedule this before filling in the page, just so it's on the calendar.
* Create the meeting on the "Incident Post-Mortem Meetings" shared calendar. * Create the meeting on the "Incident Post-Mortem Meetings" shared calendar.
@ -26,12 +26,12 @@ Once you've been designated as the owner of a post-mortem, you should start upda
1. Begin populating the page with all of the information you have. 1. Begin populating the page with all of the information you have.
* The timeline should be the main focus to begin with. * The timeline should be the main focus to begin with.
* The timeline should include important changes in status/impact, and also key actions taken by responders. * The timeline should include important changes in status/impact, and also key actions taken by responders.
* You should mark the start of the incident in red, and the resolution in green (for when we went into/out of SEV). * You should mark the start of the incident in red, and the resolution in green (for when we went into/out of SR/IN +major).
* Go through the history in Slack to identify the responders, and add them to the page. * Go through the history in DoIT and Slack to identify the responders, and add them to the page.
* Identify the Incident Commander and Scribe in this list. * Identify the Team Leader and Scribe in this list.
1. Populate the page with more detailed information. 1. Populate the page with more detailed information.
* For each item in the timeline, identify a metric, or some third-party page where the data came from. This could be a link to a Datadog graph, a SumoLogic search, a Tweet, etc. Anything which shows the data point you're trying to illustrate in the timeline. * For each item in the timeline, identify a metric, or some third-party page where the data came from. This could be a link to a Check_MK graph, a logwatch search, a Tweet, etc. Anything which shows the data point you're trying to illustrate in the timeline.
1. Perform an analysis of the incident. 1. Perform an analysis of the incident.
* Capture all available data regarding the incident. What caused it, how many customers were affected, etc. * Capture all available data regarding the incident. What caused it, how many customers were affected, etc.
@ -39,13 +39,13 @@ Once you've been designated as the owner of a post-mortem, you should start upda
* Capture the impact to customers (generally in terms of event submission, delayed processing, and slow notification delivery) * Capture the impact to customers (generally in terms of event submission, delayed processing, and slow notification delivery)
* Identify the underlying cause of the incident (What happened, and why did it happen). * Identify the underlying cause of the incident (What happened, and why did it happen).
1. Create any followup action JIRA tickets (or note down topics for discussion if we need to decide on a direction to go before creating tickets), 1. Create any followup action DoIT cards (or note down topics for discussion if we need to decide on a direction to go before creating tickets),
* Go through the history in Slack to identify any TODO items. * Go through the history in DoIT, Slack to identify any TODO items.
* Label all tickets with their severity level and date tags. * Label all tickets with their severity level and date tags.
* Any actions which can reduce re-occurrence of the incident. * Any actions which can reduce re-occurrence of the incident.
* (There may be some trade-off here, and that's fine. Sometimes the ROI isn't worth the effort that would go into it). * (There may be some trade-off here, and that's fine. Sometimes the ROI isn't worth the effort that would go into it).
* Identify any actions which can make our incident response process better. * Identify any actions which can make our incident response process better.
* Be careful with creating too many tickets. Generally we only want to create things that are P0/P1's. Things that absolutely should be dealt with. * Be careful with creating too many cards. Generally we only want to create things that are of top priority. Things that absolutely should be dealt with.
1. Write the external message that will be sent to customers. This will be reviewed during the post-mortem meeting before it is sent out. 1. Write the external message that will be sent to customers. This will be reviewed during the post-mortem meeting before it is sent out.
* Avoid using the word "outage" unless it really was a full outage, use the word "incident" instead. Customers generally see "outage" and assume everything was down, when in reality it was likely just some alerts delivered outside of SLA. * Avoid using the word "outage" unless it really was a full outage, use the word "incident" instead. Customers generally see "outage" and assume everything was down, when in reality it was likely just some alerts delivered outside of SLA.
@ -57,18 +57,18 @@ These meetings should generally last 15-30 minutes, and are intended to be a wra
You should invite the following people to the post-mortem meeting, You should invite the following people to the post-mortem meeting,
* Always * Always
* The incident commander. * The team leader.
* Service owners involved in the incident. * Service owners involved in the incident.
* Key engineer(s)/responders involved in the incident. * Key engineer(s)/responders involved in the incident.
* Optional * Optional
* Customer liaison. (Only SEV-1 incidents) * Customer liaison. (Only SR/IN +major incidents)
A general agenda for the meeting would be something like, A general agenda for the meeting would be something like,
1. Recap the timeline, to make sure everyone agrees and is on the same page. 1. Recap the timeline, to make sure everyone agrees and is on the same page.
1. Recap important points, and any unusual items. 1. Recap important points, and any unusual items.
1. Discuss how the problem could've been caught. 1. Discuss how the problem could've been caught.
* Did it show up in canary? * Did it send any weak signals?
* Could it have been caught in tests, or loadtest environment? * Could it have been caught in tests, or loadtest environment?
1. Discuss customer impact. Any comments from customers, etc. 1. Discuss customer impact. Any comments from customers, etc.
1. Review action items that have been created, discuss if appropriate, or if more are needed, etc. 1. Review action items that have been created, discuss if appropriate, or if more are needed, etc.