idnits 2.17.1 draft-clemm-netconf-push-smart-filters-ps-00.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (October 18, 2017) is 2380 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 6536 (Obsoleted by RFC 8341) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Clemm 3 Internet-Draft Futurewei Technologies, Inc. 4 Intended status: Informational E. Voit 5 Expires: April 21, 2018 Cisco Systems 6 X. Liu 7 Jabil 8 I. Bryskin 9 T. Zhou 10 G. Zheng 11 Huawei 12 H. Birkholz 13 Fraunhofer SIT 14 October 18, 2017 16 Smart filters for Push Updates - Problem Statement 17 draft-clemm-netconf-push-smart-filters-ps-00 19 Abstract 21 This document defines a problem statement for Smart Filters for Push 22 Updates. Smart Filters for Push Updates (referred to simply as 23 "Smart Filters" in the context of this document) allows to filter 24 push updates based on values of pushed objects and/or state, such as 25 previous updates. Smart Filters provide an important building block 26 for service assurance and network automation. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on April 21, 2018. 45 Copyright Notice 47 Copyright (c) 2017 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 63 2. Key Words . . . . . . . . . . . . . . . . . . . . . . . . . . 3 64 3. Definitions and Acronyms . . . . . . . . . . . . . . . . . . 3 65 4. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 3 66 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6 67 6. Security Considerations . . . . . . . . . . . . . . . . . . . 6 68 7. Normative References . . . . . . . . . . . . . . . . . . . . 6 69 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 7 71 1. Introduction 73 YANG-Push [yang-push] allows client applications to subscribe to 74 continuous datastore updates without needing to poll. YANG-Push 75 subscriptions allow client applications to select which datanodes are 76 of interest. For this purpose, filters that act as node selectors 77 are offered. However, what is currently not supported are filters 78 that filter updates based on values, such as sending updates only 79 when the value falls within a certain range. Also not supported are 80 filters that would require additional state, such as sending updates 81 only when the value exceeds a certain threshold for the first time 82 but not again until the threshold is cleared. We refer to such 83 filters as "smart filters", with further subcategories of "smart 84 stateless filters" and "smart stateful filters", respectively. 86 Smart filters involve more complex subscription and implementation 87 semantics than the simple selection filters that are currently 88 offered as part of YANG-Push. They involve post processing of 89 updates that goes beyond basic update generation for polling 90 avoidance and place additional intelligence at the server. Because 91 of this, smart filter functionality was not included in the YANG-Push 92 specification, although it was recognized that YANG-Push could be 93 extended to include such functionality if needed. This is the 94 purpose of this specification. 96 Smart filters facilitate service assurance, because they allow client 97 applications to focus on "outliers" and updates that signify 98 exceptions and conditions of interest have the biggest operational 99 significance. They save network resources by avoiding the need to 100 stream updates that would be discarded anyway, and allow applications 101 to scale better since larger networks imply a larger amount of smart 102 filtering operations delegated away from the application to the 103 network. Smart filters also facilitate network automation as they 104 constitute an important ingredient to specify triggers for automated 105 actions. 107 2. Key Words 109 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 110 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 111 "OPTIONAL" in this document are to be interpreted as described in BCP 112 14 [RFC2119] [RFC8174] when, and only when, they appear in all 113 capitals, as shown here. 115 3. Definitions and Acronyms 117 Smart Filter: A filter that involves some processing, such as 118 comparing values or differentiating behavior depending on state. 120 TCA: Threshold Crossing Alert. 122 YANG-Push: A server capability that allows client applications to 123 subscribe to network management datastore updates. 125 4. Problem Statement 127 YANG-Push provides client applications with the ability to subscribe 128 to continuous updates from network management datastores, obviating 129 the need to perform polling and resulting in more robust and 130 efficient applications. However, many applications do not require 131 every update, only updates that are of certain interest. 133 For example, an update concerning interface utilization may be only 134 needed when a certain utilization level is breached. Sending 135 continuous updates when utilization is low might divert processing 136 resources away from updates regarding interfaces whose utilization 137 level may reach a critical point that requires attention. Doing so 138 will require a filter based on an object value. Even sending 139 continuous updates when utilization is high may be too much and 140 counterproductive. It may be sufficient to send an update when a 141 threshold is breached to raise a flag of attention, but then not to 142 continue sending updates while the condition still persists but 143 simply let the client application know when the threshold is cleared. 144 This behavior cannot be accomplished simply by a value-based filter, 145 but requires additional state to be maintained (so that the server 146 has a memory whether or not the condition of a breached threshold has 147 already been reported in prior update cycles). 149 What is needed are "Smart Filters" that provide the ability to apply 150 filters based on object values, possibly also state. Smart Filters 151 are useful for Service Assurance applications that need to monitor 152 operational data for values that fall outside normal operational 153 ranges. They are also useful for network automation, in which 154 automated actions are automatically triggered based on when certain 155 events in the network occur while certain conditions hold. A YANG- 156 Push subscription with a smart filter can in effect act as a source 157 for such events. Combined with an optional check for a condition 158 when an event is observed, this can serve as the basis of action 159 triggers. 161 Of course, it is possible to conceive filters that are very smart and 162 powerful yet also very complex. While filters as defined in YANG- 163 Push may be a tad too simple for the applications envisioned here, it 164 is important to keep filters still simple enough to ensure broad 165 implementation and support by networking devices. The smart filters 166 defined in this effort intend to apply the "90/10" rule, aiming at 167 the sweet spot that addresses 90% of use cases and deployment 168 scenarios that can be addressed using 10% of the complexity. Where 169 those filters are not sufficient, additional filters can be 170 introduced outside this document. 172 It is proposed that Smart Filters for Push Updates will provide 173 support for the following features: 175 o Support for smart filter extensions to YANG-Push subscriptions. 176 The targeted model takes a "base" YANG-Push subscription and 177 subjects updates to an additional filtering stage that is based on 178 an object's value. 180 * Filters that match or compare the object value against a fixed 181 term or expression. 183 * Filters that match or compare an object value against the value 184 of another object. (This feature is useful particularly in 185 conjunction with possible extensions that allow to compute 186 aggregates. Such an ability opens the possibility to compare 187 an object's current value to its mean, for example.) 189 o Support for refined on-change update semantics that allow client 190 to distinguish whether object values were omitted or included 191 because the object was created or deleted, or because the object's 192 value fell outside filter range. 194 o Support for selected stateful filters: 196 * This includes specifically support for generalized "threshold 197 crossing alert" filters, or filters that provide an update only 198 when an object's value passes a filter for the first time, and 199 not again until the object's value passes a counter filter. In 200 effect, the support involves attaching filter and counter 201 filter to an object, including a switch at the object 202 indicating which filter is in effect, and providing a 203 distinction in the update which filter (e.g. onset of clear) 204 was applied. 206 * It may include additional filters, such a "recent high water 207 mark" filters that allow to specify a time horizon until the 208 current high water mark clears. A recent high water mark 209 filter sends an update to an object only if its new value is 210 greater than the last value that had been previously reported. 212 In order to constrain complexity, it is proposed that the following 213 items will be outside the scope, subject to discussion by the Working 214 Group: 216 o Filters that involve freely programmable logic. 218 o Filters that aggregate or otherwise process information over time. 219 An example would be filters that compute an aggregate over a time 220 series of data, for example, an object's average or top percentile 221 value. (One way in which this can be accomplished is by defining 222 a separate YANG module that allows to specify aggregates 223 independent of any filtering, then asking users to subscribe to 224 updates of the aggregate objects and applying filters there. The 225 definition of such an aggregation module goes beyond the scope of 226 the work defined here.) 228 o Filters that aggregate object's values with those of other 229 objects, such as the maximum or average from objects over a list, 230 or that operate on a function of other objects. An example would 231 be an object for interface utilization that gets computed from 232 objects for interface speed and interface packet rate, with the 233 packet rate object itself potentially computed from counter 234 snapshots that are taken at different times. (One way in which 235 this can be accomplished is by defining a separate YANG module 236 that allows to define objects that compute such functions akin to 237 the Expression MIB [RFC2982], then asking users to subscribe to 238 updates of those objects and applying filters there.) 240 5. IANA Considerations 242 Not applicable 244 6. Security Considerations 246 The application of smart filters requires a certain amount of 247 processing resources at the server. An attacker could attempt to 248 attack a server by creating YANG-push subscriptions with a large 249 number of complex smart filters in an attempt to diminish server 250 resources. Server implementations can guard against such scenarios 251 in several ways. For one, they can implement NACM [RFC6536] in order 252 to require proper authorization for requests to be made. Second, 253 server implementations can reject requests made for a a larger number 254 of smart filters than the implementation can reasonably sustain. 256 7. Normative References 258 [notif-sub] 259 Voit, E., Clemm, A., Gonzalez Prieto, A., Nilsen-Nygaard, 260 E., and A. Tripathy, "Custom subscription to event 261 notifications", July 2017, 262 . 265 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 266 Requirement Levels", BCP 14, RFC 2119, 267 DOI 10.17487/RFC2119, March 1997, 268 . 270 [RFC2982] Kavasseri, R., Ed., "Distributed Management Expression 271 MIB", RFC 2982, DOI 10.17487/RFC2982, October 2000, 272 . 274 [RFC6536] Bierman, A. and M. Bjorklund, "Network Configuration 275 Protocol (NETCONF) Access Control Model", RFC 6536, 276 DOI 10.17487/RFC6536, March 2012, 277 . 279 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 280 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 281 May 2017, . 283 [yang-push] 284 Clemm, A., Voit, E., Gonzalez Prieto, A., Tripathy, A., 285 Nilsen-Nygaard, E., Bierman, A., and B. Lengyel, 286 "Subscribing to YANG datastore push updates", August 2017, 287 . 290 Authors' Addresses 292 Alexander Clemm 293 Futurewei Technologies, Inc. 294 2330 Central Expressway 295 Santa Clara, CA 95050 296 USA 298 Email: ludwig@clemm.org 300 Eric Voit 301 Cisco Systems 303 Email: evoit@cisco.com 305 Xufeng Liu 306 Jabil 308 Email: Xufeng_Liu@jabil.com 310 Igor Bryskin 311 Huawei 313 Email: Igor.Bryskin@huawei.com 315 Tianran Zhou 316 Huawei 318 Email: zhoutianran@huawei.com 320 Guangying Zheng 321 Huawei 323 Email: zhengguangying@huawei.com 324 Henk Birkholz 325 Fraunhofer SIT 327 Email: henk.birkholz@sit.fraunhofer.de