Method and apparatus for detecting file system corruption
View Patent ↗The system and method described herein automatically detect various corruptions in a file system and notify a system administrator of the corruption. Detailed information on the file system is collected by a probe process. If the file system is corrupt or inaccessible, the system and method marks the file system as bad, notifies the system administrator and then ceases to attempt to collect information on that system again until it has been repaired.
1. A method for detecting corrupt file systems, comprising:
identifying an operating system running on a host, the host attached to at least one file system;
performing file system probes on the at least one file system, the respective file system probes being dependent on operating system type;
testing each of the at least one file system for responsiveness by appending an open file on each said at least one file system using a multi-level time-out mechanism, wherein the multi-level time-out mechanism comprises a user selectable interval for attempting an append and a user selectable maximum number of tries to attempt a successful append; and
sending, if one of the testing of the corrupt file systems exceeds the maximum number of probe tries at the selected probe interval and the file system probes identify file corruption, a notification that a given file system is corrupt.
2. The method as recited in claim 1 , wherein if the operating system type is UNIX, the file system probe identifies corruption of the host's inode table.
3. The method as recited in claim 1 , wherein if the operating system type is Windows.TM., the file system probe identifies corruption of the host's FAT table.
4. The method as recited in claim 1 , wherein the selected probe interval is one second, and is reconfigurable for different computation loads.
5. The method as recited in claim 1 , wherein the selected number of probe tries is 300, and is reconfigurable for different computation loads.
6. The method as recited in claim 1 , wherein sending a notification further comprises:
creating an event; and
performing an action corresponding to the event, wherein the action is selected from the group consisting of actions of e-mailing a file system corruption notification, paging at least one selected individual, and posting a warning to an application process.
7. The method as recited in claim 1 , further comprising identifying corruption in a file system using operating system application program interface (API) code to retrieve file system information.
8. A file system corruption detection mechanism operating on a network of computers, comprising:
means for probing a plurality of file systems on a computer network, wherein the means for probing tests the file system for corruption, and if probing fails to identify corruption, attempts to append data to an open file on a probed file system;
means for determining whether a maximum number of append tries has been reached, wherein each append try on a file system is separated by a user selectable interval;
means for determining whether a probed file system is corrupted; and
means for notifying a user of a corrupted file system.
9. The mechanism as recited in claim 8 , wherein the maximum number of append tries is 300.
10. The mechanism as recited in claim 8 , wherein the user selectable interval is one second.
11. The mechanism as recited in claim 8 , wherein the means for notifying a user generates an event, the event corresponding to an action selected from the group of actions consisting of e-mailing a file system corruption notification, paging at least one selected individual, and posting a warning to an application process.
12. The mechanism as recited in claim 8 , wherein the means for probing is customizable for an operating system type, and wherein for an operating system type of UNIX, the means for probing identifies corruption of the host's inode table.
13. The mechanism as recited in claim 8 , wherein the means for probing is customizable for an operating system type, and wherein for an operating system type of Windows™, the probes identify corruption of the host's FAT table.
14. A file system corruption detector running on a host computer, comprising:
system independent computer code for identifying system unique identifiers for detecting file system corruption; and
computer code for running a probe process, wherein the probe process opens a file on at least one file system connected to the host computer and wherein the probe process attempts to append data to the opened file, and wherein the probe process uses a multi-level time-out mechanism to determine whether the probed file system is functioning adequately, wherein the multi-level time-out mechanism comprises a user selectable interval for attempting and append and a user selectable maximum number of times to attempt a successful append.
15. The file system corruption detector as recited in claim 14 , further comprising means for performing an action if a file system is corrupted.
16. The file system corruption detector as recited in claim 15 , wherein the performed action is selected form the group consisting of actions of e-mailing a file system corruption notification, paging at least one selected individual, and posting a warning to an application process.
17. The file system corruption detector as recited din claim 15 , wherein each file system connected to the host is probed, and wherein the performed action removes the detected corrupt file system for probed file systems until the corrupt file system has been repaired.
18. The file system corruption detector as recited in claim 17 , wherein when a corrupt file system has been repaired, it is probed by the probe process.
19. A file system corruption detector running on a host computer, comprising:
system independent computer code for identifying system unique identifiers for detecting file system corruption; and
computer code for running a probe process, wherein the probe process runs on the host computer and opens a file on a least one file system, wherein the probe process attempts to append data to the opened file, wherein the probe process uses a multi-level time-out mechanism to determine whether the probed file system is functioning adequately, and wherein the multi-level time-out mechanism comprises a user selectable interval for attempting an append and a user selectable maximum number of times to attempt a successful append.
20. The file system corruption detector as recited in claim 19 , wherein the host computer is connected to at least one system over a computer network, and the probe process accesses the at least one file system over the computer network.
21. The file system corruption detector as recited in claim 20 , further comprising means for performing an action if s file system is corrupted.
22. The file system corruption detector as recited in claim 21 , wherein the performed action is selected from the group consisting of actions of e-mailing a file system corruption notification, paging at least one selected individual, and posting a warning to an application process.
23. A computer readable medium containing computer code for detecting file system corruption on a host computer, the medium comprising:
a first section of system independent computer code for identifying system unique identifiers for detecting file system corruption; and
a second section of computer code for running a probe process, wherein the probe process opens a file on at least one file system connected to the host computer, wherein the probe process attempts to append data to the opened file, wherein the probe process uses a multi-level time-out mechanism to determine whether the probed file system is functioning adequately, and wherein the multi-level time-out mechanism comprises a user selectable interval for attempting an append and a user selectable maximum number of times to attempt a successful append.
24. The computer readable medium as recited in claim 24 , further comprising a section of computer code for performing an action if a file system is corrupted.
25. The computer readable medium as recited claim 24 , wherein the performed action is selected from the group consisting of actions of e-mailing a file system corruption notification, paging at least one selected individual, and posting a warning to an application process.