Increased replication for managed volumes
A scale-out computing cluster may include a large number of computing servers and storage devices. In order to provide high reliability, the computing cluster must be able to handle failures of individual devices. Reliability of the computing cluster may be improved by providing a standby server for each active server in the computing cluster. If any active server fails, the corresponding standby server is activated. The failed server may be brought back online or replaced, at which time the restored server becomes the standby server for the now-active original standby server. During the restoration period, if any other active server fails, the standby server for that active server is immediately activated. As a result, the recovery ability of the computing cluster is only challenged if both servers of an active/standby pair fail during the restoration period, substantially improving reliability.
1. A system comprising:
a first computing cluster located in a first geographic region, the first computing cluster comprising an active index server that accesses data from a corresponding first set of data volumes;
a first application server located in the first geographic region, the first application server hosting an application;
a second computing cluster located in a second geographic region different from the first geographic region and comprising a standby index server for the active index server, the standby index server accessing data from a corresponding second set of data volumes;
a second application server located in the second geographic region; and
a name server configured to perform operations comprising:
detecting a failure of the active index server; and
in response to the detected failure, updating a name table to redirect future data access requests from the failed active index server to the standby index server for the failed active index server;
wherein the application:
detects that the name table was updated to direct future access requests to the standby index server in the second geographic region; and
in response to the detecting that the name table was updated to direct future access requests to the second computing cluster, transfers execution of itself from the first application server to the second application server.
2. The system of claim 1 , wherein the active index server is configured to perform operations comprising:
receiving data access requests;
in response to each of the data access requests:
accessing data from the corresponding first set of data volumes; and
forwarding the data access request to the standby index server.
3. The system of claim 1 , wherein the name server is further configured to perform operations comprising:
detecting a resumption of service of the failed active index server; and
in response to the detected resumption of service, updating the name table to use the failed active index server as a standby index server.
4. The system of claim 1 , further comprising:
a second standby index server for the active index server, the second standby index server accessing data from a corresponding third set of data volumes;
wherein the name server is further configured to perform operations comprising:
in response to detecting a failure of the standby index server for the failed active index server, updating the name table to direct future access requests from the standby index server for the failed active index server to the second standby index server for the failed active index server.
5. The system of claim 1 , wherein after the application is transferred to the second application server, the application accesses data by communicating within the second geographic region with the standby index server.
6. A non-transitory computer-readable medium that stores instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
accessing, by an application, an index server that accesses data from a corresponding first set of data volumes, the index server being located in a first geographic region, the one or more processors also being located in the first geographic region;
detecting that a name table was updated to direct future access requests to a second index server located in a second geographic region different from the first geographic region; and
in response to the detecting that the name table was updated to direct future access requests to the second index server, transferring execution of the application to an application server located in the second geographic region.
7. The non-transitory computer-readable medium of claim 6 , wherein after the application is transferred to the application server in the second geographic region, the application accesses data by communicating within the second geographic region with the second index server.
8. A method comprising:
detecting, by one or more processors of a name server, a failure of an active index server located in a first geographic region, the active index server accessing data from a corresponding first set of data volumes, the active index server having a corresponding standby index server located in a second geographic region, the standby index server accessing data from a corresponding second set of data volumes;
in response to the detected failure, updating a name table to redirect future data access requests from the failed active index server to the standby index server for the failed active index server;
detecting, by an application that runs in the first geographic region, that the name table was updated to direct future access requests to the standby index server located in the second geographic region different from the first geographic region; and
in response to the detecting that the name table was updated to direct future access requests to the standby index server, transferring execution of the application to an application server located in the second geographic region.
9. The method of claim 8 , further comprising:
detecting a resumption of service of the failed active index server; and
in response to the detected resumption of service, updating the name table to use the failed active index server as a standby index server.
10. The method of claim 8 , further comprising:
in response to detecting a failure of the standby index server for the failed active index server, updating the name table to direct future access requests from the standby index server for the failed active index server to a second standby index server for the failed active index server.
11. The method of claim 8 , further comprising:
receiving, by the active index server, data access requests; and
in response to each of the data access requests:
accessing data from the corresponding first set of data volumes; and
forwarding the data access request to the standby index server for the active index server.
12. The method of claim 8 , wherein the standby index server is a synchronous standby for the active index server.
13. The method of claim 8 , wherein after the application is transferred to the application server in the second geographic region, the application accesses data by communicating within the second geographic region with the standby index server.