Off-peak patching for enterprise stability
In one example, a server farm patching system may wait until fewer users are accessing a server farm to apply a patch code set to a server application executed by a server at the server farm. The server farm patching system may identify an off-peak usage time range for a server farm describing when the server farm has an activity level below an activity threshold. The server farm patching system may apply a patch code set at an off-peak usage patching aggressiveness level indicating an off-peak upper bound percentage of servers in the server farm receiving the patch code set when within the off-peak usage time range. The server farm patching system may apply the patch code set at a peak usage patching aggressiveness level indicating a peak upper bound percentage of servers in the server farm receiving the patch code set when outside the off-peak usage time range.
1. A server farm patching system comprising:
a processor;
a memory in communication with the processor, the memory storing instructions that, when executed by the processor, causes the server farm patching system to identify an off-peak usage time range for a server farm having a plurality of servers, the off-peak usage time range describing when the server farm has an activity level below an activity threshold; and
a data interface configured to apply a patch code set to one of the plurality of servers in the server farm;
wherein the data interface is configured such that:
the patch code set is applied at an off-peak usage aggressiveness level when the server farm is operating within the off-peak usage time range, the off-peak usage aggressiveness level indicating an off-peak upper bound percentage of servers in the server farm receiving the patch code set when within the off-peak usage time range, and
the patch code set is applied at a peak usage aggressiveness level when the server farm is operating outside the off-peak usage time range, the peak usage aggressiveness level indicating a peak upper bound percentage of servers in the server farm receiving the patch code set when outside the off-peak usage time range.
2. The server farm patching system of claim 1 ,
wherein the instructions further cause the server farm patching system to determine a geographic location for the server farm to identify the off-peak usage time range.
3. The server farm patching system of claim 1 , wherein the memory is configured to define the off-peak usage time range in a network container object describing the server farm.
4. The server farm patching system of claim 1 , wherein the memory is configured to designate the server farm as a single tenant server farm dedicated to a single client.
5. The server farm patching system of claim 1 , wherein the memory is configured to define a tenant mandated off-peak usage time range in a network container object describing a single tenant server farm.
6. The server farm patching system of claim 1 , wherein the data interface is further configured to apply the patch code set at a pre-set shoulder aggressiveness level to the server farm proximate to an off-peak usage time range bound.
7. The server farm patching system of claim 1 ,
wherein the instructions further cause the server farm patching system to override the off-peak usage aggressiveness level upon receiving an administrator override.
8. The server farm patching system of claim 1 , wherein the memory is configured to associate a geographic region for the server farm with the patch code set for the server farm.
9. The server farm patching system of claim 1 ,
wherein the instructions further cause the server farm patching system to validate the patch code set based on an execution of the patch code set in an execution environment for a server application incorporating the patch code set.
10. The server farm patching system of claim 1 ,
wherein the instructions further cause the server farm patching system to rate the patch code set based on at least one of a build age and a regression history for the patch code set.
11. The server farm patching system of claim 1 , wherein the data interface is further configured to place a patching bid for the patch code set for the server farm requesting to apply the patch code set.
12. The server farm patching system of claim 1 ,
wherein the instructions further cause the server farm patching system to assign a farm temperature to the server farm describing a server characteristic used to determine when to deploy the patch code set to the server farm.
13. The server farm patching of claim 1 , further comprising:
wherein the instructions further cause the server farm patching system to schedule a deployment of the patch code set based on a farm temperature for the server farm.
14. The server farm patching system of claim 1 , wherein the memory is configured to mark the patch code set upon a regression occurrence indicating that the patch code set had to be removed to revert server application to an earlier version.
15. A computing device comprising:
a processor;
a memory in communication with the processor, the memory storing a series of instructions that, when executed by the processor, cause the computing device to:
apply a patch code set to a server farm,
analyze a request-per-second history for the server farm to identify an off-peak usage time range for the sever farm describing when the server farm has an activity level below an activity threshold;
wherein:
the patch code set is applied at an off-peak usage aggressiveness level when the server farm is operating within the off-peak usage time range, the off-peak usage aggressiveness level indicating an off-peak upper bound percentage of servers in the server farm receiving the patch code set when within the off-peak usage time range, and
the patch code set is applied at a peak usage aggressiveness level when the server farm is operating outside the off-peak usage time range, the peak usage aggressiveness level indicating a peak upper bound percentage of servers in the server farm receiving the patch code set when outside the off-peak usage time range.
16. The computing device of claim 15 , wherein the instructions further cause the computing device to adjust the off-peak usage time range based on a current request-per-second report for the server farm.
17. The computing device of claim 15 , wherein the instructions further cause the computing device to adjust an aggressiveness level based on a proximity to an off-peak usage time range bound.
18. The computing device of claim 15 , wherein the instructions further cause the computing device to adjust an aggressiveness level based on a current request-per-second report.
19. A method, comprising:
identifying an off-peak usage time range for a server farm having a plurality of servers, the off-peak usage time range describing when the server farm has an activity level below an activity threshold;
applying a patch code set to one of the plurality of servers in the server farm;
wherein:
the patch code set is applied at an off-peak usage aggressiveness level when the server farm is operating within the off-peak usage time range, the off-peak usage aggressiveness level indicating of an off-peak upper bound percentage of servers in the servers farm receiving the patch code set when within the off-peak usage time range, and
the patch code set is applied at a peak usage aggressiveness level when the server farm is operating outside the off-peak usage time range, the peak usage aggressiveness level indicating a peak upper bound percentage of servers in the server farm receiving the patch code when outside the off-peak usage time range.
20. The method of claim 19 , further comprising
placing a patching bid for the patch code set for the server farm, the patching bid requesting to apply the patch code set.