Hi,

According to all docs I've read, and some code, the following settings seem to have the same effect:
1. Setting features = "1 queue_if_no_path"
2. Setting no_path_retry = "queue"

Now that's confusing, especially when having conflicting configuration, e.g., from hwtable.c:
/* IBM DS4100 */
.vendor = "IBM",
.product = "1724",
.features = "1 queue_if_no_path",
.no_path_retry = 300,

It's not clear enough what would be the resulting behavior of this configuration.

Did I get it right? Is there a reason for keeping the features = "1 queue_if_no_path" option available in multipath-tools?

Thanks

Regards

- Oren

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel

Hi,

Oren Held wrote:
> Hi,
>
> According to all docs I've read, and some code, the following settings seem to have the same effect:
> 1. Setting features = "1 queue_if_no_path"
> 2. Setting no_path_retry = "queue"
>
> Now that's confusing, especially when having conflicting configuration, e.g., from hwtable.c:
> /* IBM DS4100 */
> .vendor = "IBM",
> .product = "1724",
> .features = "1 queue_if_no_path",
> .no_path_retry = 300,
>
> It's not clear enough what would be the resulting behavior of this configuration.
>
> Did I get it right? Is there a reason for keeping the features = "1 queue_if_no_path" option available in multipath-tools?

See http://marc.info/?l=dm-devel&m=124572577028515&w=2

Thanks,
Kiyoshi Ueda


> Thanks
>
> Regards
>
> - Oren

--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel