Determining the applicability of patches for installation on a computer system

A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.

1. Field of the Invention

The present invention relates generally to techniques for maintaining programming systems, and more particularly, to methods for determining the applicability of patches for installation on a computer system.

2. Description of the Related Art

When an operating system, such as Hewlett-Packard's version of UNIX “HP-UX,” receives new program files that are to be added to a given system, the files are delivered gathered into filesets having names, such as FS 1 , FS 2 , and so on. These filesets are installed on a given system by a process that unpacks and, possibly, uncompresses the files and places them onto the hard disk drive of that system. As in shown in FIG. 1 , each fileset can contain a small or large number of files. The FILESET FS 1 is shown containing the files FILE A, FILE B, . . . and FILE F. Likewise, the FILESET FS 2 is shown containing the files FILE J, FILE K, . . . and FILE P. Of course, a fileset typically contains many more files than this. Some of these files would be program files, some would be data files, some would be graphic image and multimedia files, depending upon the particular nature of the system and the particular nature of the programming system being installed.

Patches, or corrected/updated sets of files, are also delivered to a system as collections of filesets. In the HP-UX system, it is customary that the filesets in a patch have the same names as the installed filesets. A patch fileset contains updated versions of some (possibly all) of the files in the system fileset having the same name. A given patch PATCH_ 5 contains new features and fixes or repairs for specific defects. Descriptions of the new features and of the repaired defects are contained in a text file that is maintained in a central database for each patch and that is searchable for words and phrases. Accordingly, a systems administrator may search through the patch text file database and locate patches that repair particular defects or add particular features.

Over time, a first patch may be replaced by a second patch which contains all the fixes and new features of the first patch plus additional changes. These additional changes are called incremental fixes. The new patch then SUPERSEDES the previous patch. With reference to FIG. 4 , the PATCH_ 4 at the root of the patch tree 40 supersedes all of the three patches to the left in this simple linear search tree. Historically, the first patch created was PATCH_ 1 . It was superceded by PATCH 2 , which was later superceded by PATCH 3 , and that patch was later superceded by PATCH_ 4 which now resides at the root of the patch tree 40 .

In some situations, as illustrated in FIG. 3 at 30 and also in FIG. 5 at 50 , two or more patches will be replaced by a single patch. Thus, PATCH_ 6 SUPERSEDES both the patches PATCH_ 5 and PATCH_ 8 . This is represented in the search tree by PATCH_ 6 forming the root of a sub-tree having the two branches PATCH_ 5 and PATCH_ 8 . Referring now to FIG. 5 , the same patch tree shown in FIG. 3 is shown at a later point in time. At some point in time, a new patch PATCH_ 9 was added which was not part of the original patch search tree but which initially formed a single isolated patch search tree having only one patch element. Then a new patch PATCH_ 7 was created which combined all of the updates and changes contained in the patches 5 , 6 , 8 , and 9 . Even later on, PATCH_ 7 was superceded by a new patch PATCH_ 10 , thus forming the patch tree 50 shown in FIG. 5 . The root patch in the patch tree 50 is the PATCH_ 10 . That patch and PATCH_ 7 form the trunk of this searchable patch tree, which then branches into two branches, one containing PATCH_ 9 and another containing PATCH_ 6 ; and the PATCH_ 6 branch of the tree then branches again into the two patches PATCH_ 5 and PATCH_ 8 . As can be seen, a patch tree can become quite elaborate over time as many patches are combined into a smaller number of newer patches. When placed into a patch tree database, as shown in FIG. 2 , a patch tree can be searched in an automated manner.

Patch applications are designed to identify, analyze, and deliver patches to customers. A patch is applicable to a system if at least one of the filesets contained in the patch has already been installed on the system and no successor to the patch is already installed on the system. During the identification phase, algorithms identify starting locations on patch chains and traverse the chains, analyzing the attributes of the patches attempting to identify the most appropriate patch for the customer.

Known patch applications have utilized knowledge about the system being patched. This information includes a list of installed filesets and patches, which may be used to eliminate many non-applicable patches from the search space. Without such information, the patch application is forced to assume all patches for the specified hardware version and operating system version are applicable. Moreover, when performing dependency analysis for patches given only the hardware (HW) and operating system (OS), the patch application must assume that none of the dependents are installed on the computer system. The term “dependent” patch refers to a patch which requires the additional installation of a different patch found on a separate patch tree. As a result, some dependent patches included are unnecessary because they (or one of their successors) are already installed on the customer's system. The ability to quickly and conveniently determine patch applicability is vital when developing complex patch analysis tools.

Briefly summarized, an embodiment of the invention may be found in a system and/or method which determines applicability of patches for installation on a computer system. The method includes determining whether a set of patches is stored in memory of a computer system and, if the set of patches is stored, returning the cached set of patches as an applicability result; and, if the set of patches is not stored, traversing patch chains to encounter patches in the patch chains. Traversing the patch chains begins with start patches, and encountered patches are added to the applicability result.

FIG. 1 presents the structure of a systems database that indicates which files, which filesets, and which patches are installed on each system.

FIG. 2 presents the structure of a patches database that indicates what filesets each patch corrects and which files within those filesets the patches repair or modify or both.

FIG. 3 presents the database structure of a patch tree database showing the root patch for each patch tree, the filesets that each patch tree modifies, and the non-root patches within the branches of each patch tree.

FIG. 4 presents a simple linear patch tree.

FIG. 5 presents a more complex patch tree with several branches.

FIG. 6 presents a set of four patch trees, two of which have branches.

FIG. 7 presents a patch tree.

FIG. 8 presents a flow diagram of operations to create a start patches set in accordance with an embodiment of the invention.

FIG. 9 presents a flow diagram depicting exemplary operations for determining which patches are applicable to a system in accordance with an embodiment of the invention.

A patch application is a program that guides a user or an administrator through the complex process of selecting which software patches to install, taking into consideration their stability as well as the degree of risk that this particular system may be subjected to. Patch applications need to know which patches are applicable to a system. For example, the patch application may need to know if a system is configured correctly to run application X which requires patches A, B, and C. The patch application may need to know if a system is patched according to some requirement which specifies that patches A, B, and C should be installed. Further, the patch application may need to know which of a group of patches containing critical defect fixes can be installed on this system.

FIG. 4 illustrates two exemplary patch chains or trees. A patch tree database contains these trees. A patch tree 40 includes patch_ 1 , patch_ 2 , patch_ 3 , and patch_ 4 . A patch tree 42 includes patch_ 12 , patch 13 , and patch_ 14 . Patch_ 1 is superceded by patch_ 2 . Both patch_ 12 and patch_ 14 are superceded by patch_ 13 . Patch_ 4 is the successor patch for patch_ 1 , patch_ 2 , and patch_ 3 . Table 1 lists the patches and corresponding filesets for these two exemplary patch chains. Each fileset has a fileset start patch that defines where on a patch chain a fileset is introduced. All successor patches to the fileset start patch also included that fileset.

The fileset start patches from FIG. 4 and Table 1 are as follows. The start patches for FS 1 are patch_ 1 and patch_ 13 . The start patch for FS 12 is patch_ 12 . The start patch for FS 14 is patch_ 14 . In an exemplary embodiment, a fileset start patch table is input to an algorithm and is read at application startup into a hash table to allow fast access. Advantageously, the size of the fileset start patch table is small, generally proportional to the number of filesets.

Given the table of fileset start patches, the start and successor patches that are actually applicable to an unpatched computer system (the “applicability set”) can be determined by following the patch chains in the patch tree database starting at the corresponding fileset start patch for each fileset installed on the computer system. Every patch encountered on the chain is applicable. For example, using the example patches above, a computer system containing FS 1 and FS 12 would have applicable patches patch_ 1 (and all successors patch_ 2 , patch_ 3 , patch_ 4 ) and patch_ 12 (and successor patch_ 13 ).

For a computer system that has not yet been patched but that has many filesets installed, the applicability set can be computed by following all patch chains starting with the appropriate fileset start patches, accumulating all patches as they are visited. The search tree can be improved by stopping any traversal which encounters a patch already in the set. Exemplary software routines for computing a set of all patches, computing a set of start patches, and computing a set of applicable patches are provided in an attached computer program listing appendix. The computation of the applicable patch set for a computer system that has been previously patched can be done in a similar fashion by adjusting the start set appropriately. This computation is done to eliminate parts of the patch chains which are already installed.

FIG. 7 illustrates an exemplary patch chain or tree. If the computer system is unpatched, and all of the patches in FIG. 7 are applicable; the set of start patches which define the applicability set is {patch_ 15 , patch 19 , patch_ 20 }. However, if a system has patch_ 16 installed, then the set of start patches which define the applicability set is {patch_ 17 , patch_ 20 }. If patch_ 17 is installed, the set of start patches which define the applicability set is {patch_ 20 }. If patch_ 18 is installed, the set is {}.

FIG. 8 illustrates exemplary operations in the creation of an applicability set of start patches for a patched computer by starting with a corresponding set for the unpatched system. Additional, fewer, or different operations may be performed in various processes for obtaining software patches, depending on the embodiment. In an operation 71 of an exemplary embodiment, any start patch is eliminated from the set which has an installed successor. For example, in the patch chain described with reference to FIG. 7 , if patch_ 16 is installed, patch_ 15 and patch_ 19 can be eliminated from the applicability set.

In an operation 73 , the immediate successor is added to any installed patch to the set of start patches. Again referring to FIG. 7 , if patch_ 16 is installed, its immediate successor, patch_ 17 is added to the set of start patches. In an operation 75 , any patch is removed which has a predecessor in the set. Thus, if patch_ 20 is in the set, the predecessor patch_ 18 can be removed. Exemplary software routines for these operations are provided in the attached computer program listing appendix.

Preferably, the start patches for a system are constant during analysis. As such, computation of start patches needs only occur one time during the execution of the patch application. The same can be said for the applicability set: however, this set may be too large to keep in cache.

FIG. 9 illustrates a flow diagram of exemplary operations to determine if patches are in the applicability set for a system. Additional, fewer, or different operations may be performed in various different embodiments. A system S and a set of patches, Input Patches, are input for the operations and the output is a subset of Input Patches which are applicable to S. In an operation 81 , the variables “result” and “visited” are set to empty sets. In an operation 82 , SP is set to the start patches for the patched system S. A determination is made in an operation 83 as to whether SP is empty. If SP is empty, this result is returned in an operation 84 .

If SP is not empty, an operation 85 is performed in which P is set to be a patch in SP. In an operation 86 , P is removed from SP. In an operation 87 , a query is made as to if P is null. If P is null, control returns to operation 83 . If P is not null, a query is made in an operation 88 as to if P is in “visited.” If P is in “visited,” control returns to operation 83 . If P is not in “visited,” an operation 89 is performed in which P is added to “visited.” In an operation 91 , a query is made as to if P is in “patches.” If P is in “patches,” P is added to “result” and control returns to operation 83 . If P is not in “patches,” control returns to operation 83 .

By determining patch applicability using fileset start patches as a guide, it is possible to enhance any patch application to deliver only applicable patches to customers without paying significant performance penalties. Additionally, the data which controls the patch application is then small and can be stored efficiently, allowing a solution which does not require a large or complex database.

While several embodiments of the invention have been described, it is to be understood that modifications and changes will occur to those skilled in the art to which the invention pertains. Accordingly, the claims appended to this specification are intended to define the invention precisely.

/**

*  Compute the set of all patches which includes a set of start

*  patches and all of their successors.

*/

function add_successors (Set : start_patches) : Set

{

Set result = { };

for start in start_patches

do

while (start != null)

do

if (start in result) {

//  prune the search

break;

}

else {

//  add the patch to the result

result = result + start;

}

done

done

return result;

}

/* Compute a set of start patches for a computer system */

function get_start_patches(s)

{

Set result = { }

// for each installed file set, look up the start patches

// in the hash table and add them to the result.

for fs in s.get_installed_filesets( )

do

result = result + start_patch_hash_table(fS)

done

return result;

}

/**

*  Compute the set of applicable patches for an unpatched

*  computer system.

*/

function compute_applicable_Set_unPatched (System : s) : Set

{

Set start_patches = get_start_patches(s);

return add_successors(start_patches);

}

/**

*  Compute a set of start patches for a system which has been

*  patched.

*/

function compute_start_patches_for_patched_system (System : s) : Set

{

// see above.

Set start_set = get_start_patches(s);

Set result = { };

for p in start_set

do

//search for the patch or any successor on the system

installed = s.patch_or_successor_installed(patch);

//if nothing found, add it to the result

if (installed == null)

result = result + patch;

// add the successor to the installed patch to the result.

else {

successor = installed.getSuccessor( );

if (successor != null)

result = result + successor;

}

done

// filter out any patches which have a predecessor in the result

//traverse the patch chains and keep track of visited patches

Set visited = { };

for p in result

do

// start at the successor and visit all successors of p

p = p.getSuccessor( );

while (p != null)

do

if (p in visited)

break;

else {

visited = visited + p;

p = p.getSuccessor( );

}

done

done

//remove any patch from the result which is also in the visited

// set.

Set copy = result;

result = { };

for p in copy

do

if p not in visited

result = result + p;

done

return result;

}

/** Compute the set of applicable patches for an unpatched

*  computer system.

*/

function compute_applicable_set_patched (System : s) : Set

{

Set start_patches =

compute_start_patches_for_patched_system(s);

return a dd_successors(start_patches);

}

/**

*  Given a set of patches, determine which ones are applicable

*  to a system

*/

function which_patches_are_applicable (System : s, Set: patches) : Set

{

Set applicable_set = compute_applicable set_patched(s);

Set result = { }

for s in patches

do

if (s in applicable_set)

result = result + s;

done

return result;

}

/**

*  Given a set of patches, determine which ones are applicable

*  to a system.

*/

function which_patches_are_applicable(System : s

Set: patches) : Set

{

Set result = { };

Set start_patches = compute_start_patches_for patched system(s);

Set visited = { };

// walk all patch chains starting with the start patches

// for p in start patches

do

while (p != null)

do

//stop if this chain has been traversed

if (p in Visited)

break;

else {

Visited = visited + p;

// add the encountered patch to the result

if (p in Patches)

result result + p;

}

done

done

return result;

}