Hi Steve,
On 1/13/25 11:23 AM, Steve Dickson wrote:
> Hey Anna,
>
> On 1/8/25 4:37 PM, Anna Schumaker wrote:
>> From: Anna Schumaker <anna.schumaker@xxxxxxxxxx>
>>
>> This is used to add an xprt to the switch at runtime.
>>
>> Signed-off-by: Anna Schumaker <anna.schumaker@xxxxxxxxxx>
>> ---
>> tools/rpcctl/rpcctl.man | 4 ++++
>> tools/rpcctl/rpcctl.py | 11 +++++++++++
>> 2 files changed, 15 insertions(+)
>>
>> diff --git a/tools/rpcctl/rpcctl.man b/tools/rpcctl/rpcctl.man
>> index b87ba0df41c0..2ee168c8f3c5 100644
>> --- a/tools/rpcctl/rpcctl.man
>> +++ b/tools/rpcctl/rpcctl.man
>> @@ -12,6 +12,7 @@ rpcctl \- Displays SunRPC connection information
>> .BR "rpcctl client show " "\fR[ \fB\-h \f| \fB\-\-help \fR] [ \fIXPRT \fR]"
>> .P
>> .BR "rpcctl switch" " \fR[ \fB\-h \fR| \fB\-\-help \fR] { \fBset \fR| \fBshow \fR}"
>> +.BR "rpcctl switch add-xprt" " \fR[ \fB\-h \fR| \fB\-\-help \fR] [ \fISWITCH \fR]"
>> .BR "rpcctl switch set" " \fR[ \fB\-h \fR| \fB\-\-help \fR] \fISWITCH \fBdstaddr \fINEWADDR"
>> .BR "rpcctl switch show" " \fR[ \fB\-h \fR| \fB\-\-help \fR] [ \fISWITCH \fR]"
>> .P
>> @@ -29,6 +30,9 @@ Show detailed information about the RPC clients on this system.
>> If \fICLIENT \fRwas provided, then only show information about a single RPC client.
>> .P
>> .SS rpcctl switch \fR- \fBCommands operating on groups of transports
>> +.IP "\fBadd-xprt \fISWITCH"
>> +Add an aditional transport to the \fISWITCH\fR.
>> +Note that the new transport will take its values from the "main" transport.
>> .IP "\fBset \fISWITCH \fBdstaddr \fINEWADDR"
>> Change the destination address of all transports in the \fISWITCH \fRto \fINEWADDR\fR.
>> \fINEWADDR \fRcan be an IP address, DNS name, or anything else resolvable by \fBgethostbyname\fR(3).
>> diff --git a/tools/rpcctl/rpcctl.py b/tools/rpcctl/rpcctl.py
>> index 20f90d6ca796..8722c259e909 100755
>> --- a/tools/rpcctl/rpcctl.py
>> +++ b/tools/rpcctl/rpcctl.py
>> @@ -223,6 +223,12 @@ class XprtSwitch:
>> parser.set_defaults(func=XprtSwitch.show, switch=None)
>> subparser = parser.add_subparsers()
>> + add = subparser.add_parser("add-xprt",
>> + help="Add an xprt to the switch")
>> + add.add_argument("switch", metavar="SWITCH", nargs=1,
>> + help="Name of a specific xprt switch to modify")
>> + add.set_defaults(func=XprtSwitch.add_xprt)
>> +
>> show = subparser.add_parser("show", help="Show xprt switches")
>> show.add_argument("switch", metavar="SWITCH", nargs='?',
>> help="Name of a specific switch to show")
>> @@ -246,6 +252,11 @@ class XprtSwitch:
>> return [XprtSwitch(xprt_switches / name)]
>> return [XprtSwitch(f) for f in sorted(xprt_switches.iterdir())]
>> + def add_xprt(args):
>> + """Handle the `rpcctl switch add-xprt` command."""
>> + for switch in XprtSwitch.get_by_name(args.switch[0]):
>> + write_sysfs_file(switch.path / "xprt_switch_add_xprt", "1")
>> +
>> def show(args):
>> """Handle the `rpcctl switch show` command."""
>> for switch in XprtSwitch.get_by_name(args.switch):
>
> Quick Question... Is this patch dependent on the previous
> posted kernel patches (aka NFS & SUNRPC Sysfs Improvements)
Thanks for checking! Yes, these patches (especially the last two) are dependent on the kernel patches. I'm in the process of updating these for a v2 with the kernel stuff. If you would rather me split this up into cleanup patches and then new feature patches, I can definitely do that!
Thanks,
Anna
>
> steved.
>
