[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [ipcdn] Comments on draft-ietf-ipcdn-pktc-eventmess-04.txt
Sumanth,
Regarding item #1...
Are MTAs really required to store 2^31 events? That seems like a lot. I'd be surprised that an MTA has that much memory.
If it can store 2^31 events, I suppose throttling isn't as important. Otherwise, assuming memory is much more limited, I would think the ability to throttle is desirable. Having a separate object to control local throttling seems OK to me.
greg
-----Original Message-----
From: ipcdn-bounces at ietf.org [mailto:ipcdn-bounces at ietf.org] On Behalf Of Sumanth Channabasappa
Sent: Tuesday, October 18, 2005 2:59 PM
To: ipcdn at ietf.org
Subject: RE: [ipcdn] Comments on draft-ietf-ipcdn-pktc-eventmess-04.txt
Folks,
I haven't received any comments on the clarifications requested (dated 10/07, enclosed in this email). Can I assume acceptance (status quo) on
#1 and #2?
Other than the clarifications, I am planning to check with the co-authors (and others in IPCDN) about Greg's comments regarding the MIB structure not conforming to the recommendations of RFC4181. The rest I believe, have been addressed as recommended.
Pending any other last minute comments, I plan to submit draft-05 by the end of this work-week (to meet the deadline).
- S
-----Original Message-----
From: Sumanth Channabasappa
Sent: Friday, October 07, 2005 3:21 PM
To: Eugene Nechamkin; Randy Presuhn; ipcdn at ietf.org
Subject: RE: [ipcdn] Comments on draft-ietf-ipcdn-pktc-eventmess-04.txt
Greg, Randy, Eugene,
Thanks for the comments. A few clarifications inline:
#1:
- The description makes mention of throttling events using SNMP and Syslog. What about the local log? Are events written to the local log not subject to throttling?
[s] The objective of the MIB Objects belonging to 'pktcDevEventThrottle'
was to limit the throttling of events to the network (esp. in power-up-after-a-blackout scenario), not locally.
Regarding local logging of events:
- the MTAs are currently required to store 2^31 events (in the local
log) and if we have questions regarding the size we should change the index value 'pktcDevEvLogIndex' (suggestions anyone, MTA vendors?)
- Further, if we want to throttle local logging, we should probably define a new MIB Object for local events only (since the throttling conditions would vary between network and local storage) - thoughts?
#2:
- pktcDevEventDescrText and pktcDevEventDescrClass - These objects are read-write. It seems odd to me that these objects would be writable.
Is it really intended that these are writable objects?
[s] Yes, this is intended since an operator could change the 'Description of an event' to better suit 'Operational specific values'
(via SNMP or by using the configuration file at bootstrapping time).
E.g. MSO XYZ could change a lengthy description such as 'Error in TFTP response to the MTA due to XYZ' into a shortened string such as 'TFTP-ERR:XYZ'.
>>> FYI <<<
I am working on a couple of other comments, but the following (mostly from Greg and some internal) have been incorporated as-is:
There are a number of places in the document that makes reference to "PacketCable device". Shouldn't this be "PacketCable or IPCablecom device"? Or more simply "MTA device".
The naming convention we've been using on other MIB modules in IPCDN adds "-IETF-" to the module name to distinguish it from the corresponding MIB module being developed in CableLabs. So, it this case, the MIB module name should be something like "PKTC-IETF-EVENT-MIB".
I think the ORGANIZATION clause should be IPCDN rather than CableLabs.
The DESCRIPTION clause is missing the mandatory copyright notice.
pktcDevEnvetReportStatus - The last sentence of the DESCRIPTION clause states "... Defined by PacketCable by default." Should this just PacketCable or PacketCable and IPCablecom.
pktcDevEvThrottleAdminStatus
- "A value of stopAtThreshold(3) causes event message transmission to cease at the threshold, and not resume until directed to do so." How is the device directed to resume sending event messages?
pktcDevEvThrottleInterval - The DEFVAL clause should be on a separate line for readability.
pktcDevEventDescrTable - Should the reference to "PacketCable" be "PacketCable/IPCablecom"?
pktcDevEventDescrId - "The event identifier can either be PacketCable defined or vendor-specific." Would be good to provide a reference to the PacketCable spec where the events are defined.
pktcDevEventDescrReporting - Should define what each one of the bit values mean.
pktcDevEvLogCorrelationId - "...per section 5.4.5 of [3]" Reference [3] doesn't exist in the references section.
pktcDevEvLogCorrelationId
"This MIB Object contains the correlation ID
generated by the MTA during the initiation of the
last provisioning flow, within or following which
the event occurred.
For information on the generation of correlation ids,
refer to the corresponding PacketCable/IPCableCom
Device Provisioning specifications."
[PKT-SP-EVEMIB1.5] - There is an odd character after the "PacketCable"
Reference section - There are number of reference in the body of the text that do not exist in the references section. Need to make sure all reference are cited. The ones I found are PKT-SP-PROV, PKT-SP-MIB-MTA, PKT-SP-MGCP, RFC3435, PKT-SP-CODEC, and RFC2119.
Security Considerations - Per the MIB Guidelines, need to explicit discuss all MIB objects, even if to only state that there are no security issues with the object.
- There is no MIB Object defined as 'pktcDevEventThrottle 3' (we have MIB Objects for 2 and 4)
- The Reference in the description of the MIB Object 'pktcDevEvLogCorrelationId' seems to be out of place.
_______________________________________________
IPCDN mailing list
IPCDN at ietf.org
https://www1.ietf.org/mailman/listinfo/ipcdn
_______________________________________________
IPCDN mailing list
IPCDN at ietf.org
https://www1.ietf.org/mailman/listinfo/ipcdn
_______________________________________________
IPCDN mailing list
IPCDN at ietf.org
https://www1.ietf.org/mailman/listinfo/ipcdn