i dont see any mention of this issue in the util-linux-ng archives (or i cant
search), but linux-2.6.19 and older had a bug in the /proc/swaps code where
the header would not be displayed (the first line). most people report the
issue as a sequence of swapon/swapoff calls to trigger, but for some "lucky"
reason it triggers for me all the time (no swap calls).
there are two aspects to the issue ... the first of which i think we should
handle while i'm on the fence about the second.
first, i dont think we should whine about unexpected format if the file is
empty (the default at boot). this is easy to do by putting the warning behind
a check to ferror().
second, we could detect that the first line we read isnt actually the header
but instead is a valid swap line and so needs processing. this relies on the
assumption that the first line will always be the same format. looking
quickly at older versions shows that this header has retained its exact format
since at least linux-2.2.0 and considering the concern that goes along with
/proc/ files and the ABI, it's highly unlikely it will ever change.
diff --git a/mount/swapon.c b/mount/swapon.c
index 55933b1..d89c6a9 100644
--- a/mount/swapon.c
+++ b/mount/swapon.c
@@ -135,11 +135,18 @@ read_proc_swaps(void) {
/* skip the first line */
if (!fgets(line, sizeof(line), swaps)) {
- warnx(_("%s: unexpected file format"), _PATH_PROC_SWAPS);
+ /* do not whine about an empty file */
+ if (ferror(swaps))
+ warn(_("%s: unexpected file format"), _PATH_PROC_SWAPS);
fclose(swaps);
return;
}
+ /* make sure the first line is the header */
+ if (line[0] != '\0' && strncmp(line, "Filename\t", 9))
+ goto valid_first_line;
+
while (fgets(line, sizeof(line), swaps)) {
+ valid_first_line:
/*
* Cut the line "swap_device ... more info" after device.
* This will fail with names with embedded spaces.
-mike
Attachment:
signature.asc
Description: This is a digitally signed message part.
