* [PATCH][IPVS] Don't leak sysctl tables if the scheduler registration fails
@ 2007-12-03 10:04 Pavel Emelyanov
2007-12-04 1:36 ` Simon Horman
0 siblings, 1 reply; 3+ messages in thread
From: Pavel Emelyanov @ 2007-12-03 10:04 UTC (permalink / raw)
To: Herbert Xu, David Miller
Cc: Linux Netdev List, devel, Wensong Zhang, Simon Horman,
Julian Anastasov
In case we load lblc or lblcr module we can leak some sysctl
tables if the call to register_ip_vs_scheduler() fails.
I've looked at the register_ip_vs_scheduler() code and saw, that
the only reason to fail is the name collision, so I think that
with some 3rd party schedulers this becomes a relevant issue. No?
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
---
diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
index b843a11..ad89644 100644
--- a/net/ipv4/ipvs/ip_vs_lblc.c
+++ b/net/ipv4/ipvs/ip_vs_lblc.c
@@ -580,9 +580,14 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler =
static int __init ip_vs_lblc_init(void)
{
+ int ret;
+
INIT_LIST_HEAD(&ip_vs_lblc_scheduler.n_list);
sysctl_header = register_sysctl_table(lblc_root_table);
- return register_ip_vs_scheduler(&ip_vs_lblc_scheduler);
+ ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler);
+ if (ret)
+ unregister_sysctl_table(sysctl_header);
+ return ret;
}
diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
index e5b323a..2a5ed85 100644
--- a/net/ipv4/ipvs/ip_vs_lblcr.c
+++ b/net/ipv4/ipvs/ip_vs_lblcr.c
@@ -769,9 +769,14 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler =
static int __init ip_vs_lblcr_init(void)
{
+ int ret;
+
INIT_LIST_HEAD(&ip_vs_lblcr_scheduler.n_list);
sysctl_header = register_sysctl_table(lblcr_root_table);
- return register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
+ ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
+ if (ret)
+ unregister_sysctl_table(sysctl_header);
+ return ret;
}
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH][IPVS] Don't leak sysctl tables if the scheduler registration fails
2007-12-03 10:04 [PATCH][IPVS] Don't leak sysctl tables if the scheduler registration fails Pavel Emelyanov
@ 2007-12-04 1:36 ` Simon Horman
2007-12-04 8:43 ` David Miller
0 siblings, 1 reply; 3+ messages in thread
From: Simon Horman @ 2007-12-04 1:36 UTC (permalink / raw)
To: Pavel Emelyanov
Cc: Herbert Xu, David Miller, Linux Netdev List, devel, Wensong Zhang,
Julian Anastasov
On Mon, Dec 03, 2007 at 01:04:53PM +0300, Pavel Emelyanov wrote:
> In case we load lblc or lblcr module we can leak some sysctl
> tables if the call to register_ip_vs_scheduler() fails.
This looks correct to me.
> I've looked at the register_ip_vs_scheduler() code and saw, that
> the only reason to fail is the name collision, so I think that
> with some 3rd party schedulers this becomes a relevant issue. No?
I guess so. Though presumably register_ip_vs_scheduler() could
have other modes of failure in the future.
> Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Simon Horman <horms@verge.net.au>
> ---
>
> diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
> index b843a11..ad89644 100644
> --- a/net/ipv4/ipvs/ip_vs_lblc.c
> +++ b/net/ipv4/ipvs/ip_vs_lblc.c
> @@ -580,9 +580,14 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler =
>
> static int __init ip_vs_lblc_init(void)
> {
> + int ret;
> +
> INIT_LIST_HEAD(&ip_vs_lblc_scheduler.n_list);
> sysctl_header = register_sysctl_table(lblc_root_table);
> - return register_ip_vs_scheduler(&ip_vs_lblc_scheduler);
> + ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler);
> + if (ret)
> + unregister_sysctl_table(sysctl_header);
> + return ret;
> }
>
>
> diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
> index e5b323a..2a5ed85 100644
> --- a/net/ipv4/ipvs/ip_vs_lblcr.c
> +++ b/net/ipv4/ipvs/ip_vs_lblcr.c
> @@ -769,9 +769,14 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler =
>
> static int __init ip_vs_lblcr_init(void)
> {
> + int ret;
> +
> INIT_LIST_HEAD(&ip_vs_lblcr_scheduler.n_list);
> sysctl_header = register_sysctl_table(lblcr_root_table);
> - return register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
> + ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler);
> + if (ret)
> + unregister_sysctl_table(sysctl_header);
> + return ret;
> }
>
>
--
Horms
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH][IPVS] Don't leak sysctl tables if the scheduler registration fails
2007-12-04 1:36 ` Simon Horman
@ 2007-12-04 8:43 ` David Miller
0 siblings, 0 replies; 3+ messages in thread
From: David Miller @ 2007-12-04 8:43 UTC (permalink / raw)
To: horms; +Cc: xemul, herbert, netdev, devel, wensong, ja
From: Simon Horman <horms@verge.net.au>
Date: Tue, 4 Dec 2007 10:36:48 +0900
> On Mon, Dec 03, 2007 at 01:04:53PM +0300, Pavel Emelyanov wrote:
> > In case we load lblc or lblcr module we can leak some sysctl
> > tables if the call to register_ip_vs_scheduler() fails.
>
> This looks correct to me.
>
> > I've looked at the register_ip_vs_scheduler() code and saw, that
> > the only reason to fail is the name collision, so I think that
> > with some 3rd party schedulers this becomes a relevant issue. No?
>
> I guess so. Though presumably register_ip_vs_scheduler() could
> have other modes of failure in the future.
>
> > Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
>
> Acked-by: Simon Horman <horms@verge.net.au>
Applied, thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2007-12-04 8:43 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-12-03 10:04 [PATCH][IPVS] Don't leak sysctl tables if the scheduler registration fails Pavel Emelyanov
2007-12-04 1:36 ` Simon Horman
2007-12-04 8:43 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).