Skip to contentSkip to conversation

tools/xenstore: don't close connection in xs_talkv()

PATCH In review Juergen Gross 3 participants 4 messages Mar 23, 2020

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tools/xenstore: don't close connection in xs_talkv()

In case of some errors xs_talkv() will close the connection to Xenstore. This is annoying as it is not clear to the caller in which error case the connection is still available.

Drop that implicit closing to make the interface behave in a sane and predictable way.

---
 tools/xenstore/xs.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)
tools/xenstore/xs.c +6 -8
index aa1d24b8b9..bbf3e00bed 100644
@@ -524,12 +524,14 @@ static void *xs_talkv(struct xs_handle *h, xs_transaction_t t,
 			goto fail;
 
 	ret = read_reply(h, &msg.type, len);
-	if (!ret)
-		goto fail;
 
 	mutex_unlock(&h->request_mutex);
 
 	sigaction(SIGPIPE, &oldact, NULL);
+
+	if (!ret)
+		return NULL;
+
 	if (msg.type == XS_ERROR) {
 		saved_errno = get_error(ret);
 		free(ret);
@@ -539,19 +541,15 @@ static void *xs_talkv(struct xs_handle *h, xs_transaction_t t,
 
 	if (msg.type != type) {
 		free(ret);
-		saved_errno = EBADF;
-		goto close_fd;
+		errno = EBADF;
+		return NULL;
 	}
 	return ret;
 
 fail:
-	/* We're in a bad state, so close fd. */
 	saved_errno = errno;
 	mutex_unlock(&h->request_mutex);
 	sigaction(SIGPIPE, &oldact, NULL);
-close_fd:
-	close(h->fd);
-	h->fd = -1;
 	errno = saved_errno;
 	return NULL;
 }
Signature · 7 lines
2.16.4


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Reviewed-by on v1 1/1
#1 lore raw
Quoted 8 lines · Juergen Gross

In case of some errors xs_talkv() will close the connection to Xenstore. This is annoying as it is not clear to the caller in which error case the connection is still available.

Drop that implicit closing to make the interface behave in a sane and predictable way.

This definitely does improve the cascade failure cases.

#2 lore raw
Quoted 5 lines · Andrew Cooper
Quoted 4 lines · Juergen Gross

In case of some errors xs_talkv() will close the connection to Xenstore. This is annoying as it is not clear to the caller in which error case the connection is still available.

Quoted 8 lines
Quoted 4 lines

Drop that implicit closing to make the interface behave in a sane and predictable way.

This definitely does improve the cascade failure cases.

Actually, I spoke too soon.  The EBADF goes, but the next read_message() ends up pulling junk out of the ring.

I think something else in the middle needs teaching about hard errors, and not to pull more data out of the ring.

~Andrew

#3 lore raw

Sorry to come into this late.

Andrew Cooper writes ("Re: [Xen-devel] [PATCH] tools/xenstore: don't close connection in xs_talkv()"):

Quoted 9 lines · Andrew Cooper
Quoted 8 lines · Juergen Gross
Quoted 7 lines

In case of some errors xs_talkv() will close the connection to Xenstore. This is annoying as it is not clear to the caller in which error case the connection is still available.

Drop that implicit closing to make the interface behave in a sane and predictable way.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

This definitely does improve the cascade failure cases.

Actually, I spoke too soon.  The EBADF goes, but the next read_message() ends up pulling junk out of the ring.

I'm afraid this is predictable. For most of these errors there is nothing else sensible that the client library could do. The connection has been rendered unuseable. In principle I guess it could reconnect to the socket, but this is a "should never happend" event.

EBADF is a worrying error because in many cases it means something has messed up the process's fd space, which can easily lead to really bad behaviours.

But having read the code in xenstore-ls, I see that as well as closing the fd it sets the variable containing the fd number to -1. So all future calls return EBADF.

I think this is correct.

Ian.