Windows: Work around intermittent failures in mingw_rename

We have replaced rename() with a version that can rename a file to a destination that already exists. Nevertheless, many users, the author included, observe failures in the code that are not reproducible. The theory is that the failures are due to some other process that happens to have opened the destination file briefly at the wrong moment. (And there is no way on Windows to delete or replace a file that is currently open.) The most likely candidate for such a process is a virus scanner. The failure is more often observed while there is heavy git activity (for example while the test suite is running or during a rebase operation). We work around the failure by retrying the rename operation if it failed due to ERROR_ACCESS_DENIED. The retries are delayed a bit: The first only by giving up the time slice, the next after the minimal scheduling granularity, and if more retries are needed, then we wait some non-trivial amount of time with exponential back-off. Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Johannes Sixt committed Apr 3, 2009 at 08:49 UTC 6ac6f87818edfd33fea8ce3c95b47fe00a4e4b91

1 file changed +19 -2

compat/mingw.c

+19 -2

@@ -934,7 +934,9 @@ int mingw_connect(int sockfd, struct sockaddr *sa, size_t sz)

934 #undef rename

935 int mingw_rename(const char *pold, const char *pnew)

936 {

937 - DWORD attrs;

937 + DWORD attrs, gle;

938 + int tries = 0;

939 + static const int delay[] = { 0, 1, 10, 20, 40 };

940

941 /*

942 * Try native rename() first to get errno right.

@@ -944,10 +946,12 @@ int mingw_rename(const char *pold, const char *pnew)

946 return 0;

947 if (errno != EEXIST)

948 return -1;

949 +repeat:

950 if (MoveFileEx(pold, pnew, MOVEFILE_REPLACE_EXISTING))

951 return 0;

952 /* TODO: translate more errors */

950 - if (GetLastError() == ERROR_ACCESS_DENIED &&

953 + gle = GetLastError();

954 + if (gle == ERROR_ACCESS_DENIED &&

955 (attrs = GetFileAttributes(pnew)) != INVALID_FILE_ATTRIBUTES) {

956 if (attrs & FILE_ATTRIBUTE_DIRECTORY) {

957 errno = EISDIR;

@@ -957,10 +961,23 @@ int mingw_rename(const char *pold, const char *pnew)

961 SetFileAttributes(pnew, attrs & ~FILE_ATTRIBUTE_READONLY)) {

962 if (MoveFileEx(pold, pnew, MOVEFILE_REPLACE_EXISTING))

963 return 0;

964 + gle = GetLastError();

965 /* revert file attributes on failure */

966 SetFileAttributes(pnew, attrs);

967 }

968 }

969 + if (tries < ARRAY_SIZE(delay) && gle == ERROR_ACCESS_DENIED) {

970 + /*

971 + * We assume that some other process had the source or

972 + * destination file open at the wrong moment and retry.

973 + * In order to give the other process a higher chance to

974 + * complete its operation, we give up our time slice now.

975 + * If we have to retry again, we do sleep a bit.

976 + */

977 + Sleep(delay[tries]);

978 + tries++;

979 + goto repeat;

980 + }

981 errno = EACCES;

982 return -1;

983 }