Apparatus for synchronizing operator initiated commands with a failover process in a distributed processing system

• US 6,012,150 A
• Filed: 03/27/1997
• Issued: 01/04/2000
• Est. Priority Date: 03/27/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Doxtader, John W., Reilly, Kevin J., Merritt, Laura J., LeVangia, Peter J., Rash, Nicholas P., Divirgilio, John, Bartfai, Robert F.
• Primary Examiner(s): Le, Dieu-Minh T.
• Application Number:
  US08/827,133
• Time in Patent Office: 1,013 Days
• Field of Search: 395/182.02, 395/182.05, 395/182.09, 395/182.1, 395/182.11, 395/182.13, 395/182.21, 395/183.17, 395/200.78, 395/553, 395/561, 395/595 , 395/670, 395/671, 395/672, 395/881, 371/5.4, 371/47.1, 340/825.14, 714/4, 714/7, 714/11, 714/12, 714/13, 709/248, 709/400, 710/61
• US Class Current: 714/4.11
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence

Abstract

An apparatus for synchronizing operator commands with a failover process in a distributed system having a control workstation and a plurality of nodes. One of the nodes of the distributed system is designated a primary node and one of the nodes is designated a backup node. The backup node includes a backup daemon for performing a failover process if the primary node fails such that the backup node becomes the primary node. Shell scripts send a command string to be synchronized with the operation of the backup daemon from the control workstation to the backup node. The backup daemon is then checked to determine if the backup daemon is sleeping, and, in the event the backup daemon is sleeping, commands derived from the command string are enqueued in a work queue for processing by the backup daemon. The backup daemon is then awakened such that the derived commands in the work queue are processed. In the event that the backup daemon is busy, commands derived from the command string are failed, thereby synchronizing the derived commands with the processing of the backup daemon.

Claims

1. An apparatus for preventing duplicate work from being done while performing operator commands during a failover process in a distributed system having a control workstation and a plurality of nodes, one of said nodes being designated a primary node and one of said nodes being designated a backup node, said backup node having a backup daemon which may be either asleep or awake, the apparatus comprising:

means for sending a command string from the control work station to the backup node, said command string for performing the same work as said backup daemon;

means for checking said backup daemon in said backup node to determine if said backup daemon is either awake or sleeping;

means for inserting subcommands derived from said command string into a queue in the event said backup daemon is sleeping, wherein said subcommands are enqueued for processing by said backup daemon;

means for waking up said backup daemon in the event said backup daemon is asleep, such that said backup daemon will process said derived subcommands in said queue; and

means for ending subcommands derived from said command string in the event said backup daemon is not sleeping, thereby preventing work from being done by said derived subcommands which would be duplicative of work done during the processing of said backup daemon.

2. The apparatus of claim 1 wherein said means for checking said backup daemon comprises means for checking the contents of an event word to determine if said event word contains either a null value indicating said backup daemon is busy, or a process identification value indicating said backup daemon is sleeping.

3. The apparatus of claim 2 wherein means for ending subcommands comprises means for returning a failure indication to the control workstation, thereby indicating said backup daemon is busy.

4. An apparatus for preventing duplicate work from being done while performing operator commands during a failover process in a distributed system having a control workstation and a plurality of nodes, one of said nodes being designated a primary node and one of said nodes being designated a backup node, said backup node having a backup daemon which may be either asleep or awake, the apparatus comprising:

means for sending a command string from the control work station to the backup node, said command string for performing the same work as said backup daemon;

means for checking said backup daemon in said backup node to determine if said backup daemon is either awake or sleeping;

means for inserting subcommands derived from said command string into a queue in the event said backup daemon is sleeping, wherein said subcommands are enqueued for processing by said backup daemon;

means for waking up said backup daemon in the event said backup daemon is asleep, such that said backup daemon will process said derived subcommands in said queue; and

means for ending subcommands derived from said command string in the event said backup daemon is not sleeping, thereby preventing work from being done by said derived subcommands which would be duplicative of work done during the processing of said backup daemon;

wherein said means for checking said backup daemon comprises means for checking the contents of an event word to determine if said event word contains either a null value indicating said backup daemon is busy, or a process identification value indicating said backup daemon is sleeping;

wherein said means for waking up said backup daemon comprises placing a null value in said event word indicating said backup daemon is busy.

5. The apparatus of claim 4 wherein said means for inserting subcommands into said queue comprises means for enqueueing a RESET-TO-SECONDARY subcommand into said queue.

6. The apparatus of claim 5 further comprising:

means for returning a success indication to said workstation subsequent to waking up said backup daemon, and

means for issuing from said workstation to a specified node, a BECOME-PRIMARY subcommand responsive to said success indication.

7. An apparatus comprising:

a control workstation;

a plurality of nodes, one of said nodes being designated a primary node and one of said nodes being designated a backup node, said backup node having a user space for storing user programs and a kernel space for storing kernel programs;

a backup daemon in the user space of said backup node;

shell scripts in said control workstation sending a command string from said control workstation to said backup node, said command string to be synchronized with the operation of said backup daemon;

a user module in the user space of said backup node and communicating with said shell scripts;

a kernel module in said kernel space communicating with said user module;

a daemon checking routine in said kernel module checking said backup daemon in said backup node to determine if said backup daemon is sleeping;

a work queue in said kernel space for storing work for said backup daemon;

a command inserting routine in said kernel module, said command inserting routine inserting subcommands derived from said command string into said work queue in the event said backup daemon is sleeping, wherein said commands are enqueued for processing by said backup daemon;

a wake up routine in said kernel module waking up said backup daemon in the event said backup daemon is asleep such that said backup daemon will process said derived subcommands in said queue; and

a fail routine in said kernel module ending subcommands derived from said command string in the event said backup daemon is not sleeping, thereby synchronizing said derived subcommands with the processing of said backup daemon.

8. The apparatus of claim 7 further comprising:

an event word in said kernel space containing a value indicative if said backup daemon is sleeping, said daemon checking routine checking the contents of an event word to determine if said event word contains either a null value indicating said backup daemon is busy, or a process identification value indicating said backup daemon is sleeping.

9. The apparatus of claim 8 further comprising:

a failure indication routine in said fail routine, said failure indication routine returning a failure indication to said user module for forwarding to the control workstation, thereby indicating said backup daemon is busy.

10. An apparatus comprising:

a control workstation;

a plurality of nodes, one of said nodes being designated a primary node and one of said nodes being designated a backup node, said backup node having a user space for storing user programs and a kernel space for storing kernel programs;

a backup daemon in the user space of said backup node;

shell scripts in said control workstation sending a command string from said control workstation to said backup node, said command string to be synchronized with the operation of said backup daemon;

a user module in the user space of said backup node and communicating with said shell scripts;

a kernel module in said kernel space communicating with said user module;

a daemon checking routine in said kernel module checking said backup daemon in said backup node to determine if said backup daemon is sleeping;

a work queue in said kernel space for storing work for said backup daemon;

a command inserting routine in said kernel module, said command inserting routine inserting subcommands derived from said command string into said work queue in the event said backup daemon is sleeping, wherein said commands are enqueued for processing by said backup daemon;

a wake up routine in said kernel module waking up said backup daemon in the event said backup daemon is asleep such that said backup daemon will process said derived subcommands in said queue; and

a fail routine in said kernel module ending subcommands derived from said command string in the event said backup daemon is not sleeping, thereby synchronizing said derived subcommands with the processing of said backup daemon

an event word in said kernel space containing a value indicative if said backup daemon is sleeping, said daemon checking routine checking the contents of an event word to determine if said event word contains either a null value indicating said backup daemon is busy, or a process identification value indicating said backup daemon is sleeping

wherein said wake up routine places a null value in said event word indicating said backup daemon is busy.

11. The apparatus of claim 10 wherein said command inserting routine enqueues a RESET-TO-SECONDARY subcommand into said queue.

12. The apparatus of claim 11 further comprising:

a success returning routine in said kernel module, said success returning routine returning a success indication to said user module for forwarding to said control workstation subsequent to waking up said backup daemon, and

said shell scripts including scripts for issuing from said control workstation to a specified node, a BECOME-PRIMARY subcommand responsive to said success indication.