Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > Tom Lane wrote:
> >> "full access permissions" seems unhelpfully vague. Why not say
> >> "you must have both read and write access to the current directory"?
>
> > OK, I can do that, but they need execute permission in that directory
> > too to look up file names in there. Should I say execute too?
>
> I doubt it's worth worrying about. man chdir saith
>
> In order for a directory to become the current directory, a process must
> have execute (search) access to the directory.
>
> I'm not entirely certain what happens if you chdir into a directory and
> then someone revokes the bit afterwards, but I do not feel a need to
> complicate the error message to cover such a case.
OK, fixed the the attached applied patch.
--
Bruce Momjian <bruce@momjian.us> http://momjian.us
EnterpriseDB http://enterprisedb.com
+ It's impossible for everything to be true. +
diff --git a/contrib/pg_upgrade/exec.c b/contrib/pg_upgrade/exec.c
index a66aeb0..6f1c6ef 100644
--- a/contrib/pg_upgrade/exec.c
+++ b/contrib/pg_upgrade/exec.c
@@ -96,7 +96,7 @@ verify_directories(void)
if (access(".", R_OK | W_OK | X_OK) != 0)
pg_log(PG_FATAL,
- "You must have full access permissions in the current directory.\n");
+ "You must have read and write access in the current directory.\n");
prep_status("Checking old data directory (%s)", old_cluster.pgdata);
check_data_dir(old_cluster.pgdata);
