[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16114792#comment-16114792 ] ASF GitHub Bot commented on NIFI-4022: -- Github user YolandaMDavis commented on the issue: https://github.com/apache/nifi/pull/2046 @bbende thanks! > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > Fix For: 1.4.0 > > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16114763#comment-16114763 ] ASF GitHub Bot commented on NIFI-4022: -- Github user asfgit closed the pull request at: https://github.com/apache/nifi/pull/2046 > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16114761#comment-16114761 ] ASF subversion and git services commented on NIFI-4022: --- Commit afd4f9e034e70e996abcf7e31382ee5c42a72445 in nifi's branch refs/heads/master from [~YolandaMDavis] [ https://git-wip-us.apache.org/repos/asf?p=nifi.git;h=afd4f9e ] NIFI-4022 - Initial update for SASL support for cluster management in Zookeeper NIFI-4022 - adding sasl documentation update and update to test This closes #2046. Signed-off-by: Bryan Bende> Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16114755#comment-16114755 ] ASF GitHub Bot commented on NIFI-4022: -- Github user bbende commented on the issue: https://github.com/apache/nifi/pull/2046 +1 Finally got my environment setup correctly and was able to verify the ACLs get created appropriately with this change, will merge to master, thanks! > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16113136#comment-16113136 ] ASF GitHub Bot commented on NIFI-4022: -- Github user bbende commented on the issue: https://github.com/apache/nifi/pull/2046 Reviewing... > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16109588#comment-16109588 ] ASF GitHub Bot commented on NIFI-4022: -- GitHub user YolandaMDavis opened a pull request: https://github.com/apache/nifi/pull/2046 NIFI-4022 - Enabled SASL auth scheme/ACL support for Curator use Enhancement allows user to enable SASL based ACL's for nodes created via Curator for cluster management (e.g. leader election nodes, Cluster Coordinator/Primary Nodes). For testing would recommend the following actions: 1) Follow the updated administrator guide (included in PR as a separate commit) for enabling kerberos on Zookeeper (external or embedded) and NiFi 2)Testing with nifi nodes where principals vary across servers. For example nifi/instan...@realm.com vs nifi/instan...@realm.com. In this case the kerberos.removeHostFromPrincipal would need to be true (in both zookeeper.properties and nifi.properties) to ensure that the user will be normalized as n...@realm.com for acls. 3) Ensuring leader election scenarios work as expected with acls in place on the /nifi path (acl should be 'sasl', cdrwa and 'world', anyone r). Recommended scenario is removal of Cluster Coordinator from a cluster to ensure new coordinator is elected. Thank you for submitting a contribution to Apache NiFi. In order to streamline the review of the contribution we ask you to ensure the following steps have been taken: ### For all changes: - [x] Is there a JIRA ticket associated with this PR? Is it referenced in the commit message? - [x] Does your PR title start with NIFI- where is the JIRA number you are trying to resolve? Pay particular attention to the hyphen "-" character. - [ ] Has your PR been rebased against the latest commit within the target branch (typically master)? - [ ] Is your initial contribution a single, squashed commit? ### For code changes: - [ ] Have you ensured that the full suite of tests is executed via mvn -Pcontrib-check clean install at the root nifi folder? - [x] Have you written or updated unit tests to verify your changes? - [ ] If adding new dependencies to the code, are these dependencies licensed in a way that is compatible for inclusion under [ASF 2.0](http://www.apache.org/legal/resolved.html#category-a)? - [ ] If applicable, have you updated the LICENSE file, including the main LICENSE file under nifi-assembly? - [ ] If applicable, have you updated the NOTICE file, including the main NOTICE file found under nifi-assembly? - [ ] If adding new Properties, have you added .displayName in addition to .name (programmatic access) for each of the new properties? ### For documentation related changes: - [x] Have you ensured that format looks appropriate for the output in which it is rendered? ### Note: Please ensure that once the PR is submitted, you check travis-ci for build issues and submit an update to your PR as soon as possible. You can merge this pull request into a Git repository by running: $ git pull https://github.com/YolandaMDavis/nifi NIFI-4022 Alternatively you can review and apply these changes as the patch at: https://github.com/apache/nifi/pull/2046.patch To close this pull request, make a commit to your master/trunk branch with (at least) the following in the commit message: This closes #2046 commit 9e43229ed409527ffe3bab0b3bdb7584e64ce98e Author: Yolanda M. DavisDate: 2017-07-31T17:27:48Z NIFI-4022 - Initial update for SASL support for cluster management in Zookeeper commit 588a5ca995c46f94e893b249a787be7c8104e060 Author: Yolanda M. Davis Date: 2017-08-01T18:31:15Z NIFI-4022 - adding sasl documentation update and update to test > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16108164#comment-16108164 ] Yolanda M. Davis commented on NIFI-4022: Adding some thoughts on approach given the challenges: 1) For existing NiFi users how will we migrate Zookeeper entries from the old security scheme to the new scheme? 2) How should zNodes be reverted to open if kerberos is disabled? Suggest a utility be written to migrate existing entries to the new secured scheme as well as reverse those entries if kerberos is disabled. An example of a similar approach in Kafka can be found here (e.g. kafka https://kafka.apache.org/documentation/#zk_authz) 4) Currently users can control authentication scheme via state management configuration for processors yet not for clusters. Should we still maintain the practice of allowing schemes to be configurable for processors (with SASL being the new default)? Today if security is Zookeeper using SASL and if users select Creator Only for processors then SASL auth scheme will be used. However in my opinion there is no particular driver that stands out for processor state management to enforce Creatory Only by default when security is enabled. With cluster management at the least there should be flexibility to enable SASL acl's when secured Zookeeper environment is available. > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis >Assignee: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)
[jira] [Commented] (NIFI-4022) Use SASL Auth Scheme For Secured Zookeeper Client Interaction
[ https://issues.apache.org/jira/browse/NIFI-4022?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16058067#comment-16058067 ] Jeff Storck commented on NIFI-4022: --- Adding some JIRAs related to state migration from local to ZK and usage of ZK in a standalone node of NiFi, since the scenarios described in those JIRAs should also take into consideration the usage of SASL. > Use SASL Auth Scheme For Secured Zookeeper Client Interaction > - > > Key: NIFI-4022 > URL: https://issues.apache.org/jira/browse/NIFI-4022 > Project: Apache NiFi > Issue Type: Bug >Affects Versions: 1.2.0 >Reporter: Yolanda M. Davis > > NiFi uses Zookeeper to assist in cluster orchestration including leader > elections for Primary Node and Cluster Coordinator and to store state for > various processors (such as MonitorActivity). In secured Zookeeper > environments (supported by SASL + Kerberos) NiFi should protect the zNodes it > creates to prevent users or hosts, outside of a NiFi cluster, from accessing > or modifying entries. In its current implementation security can be enforced > for processors that store state information in Zookeeper, however zNodes used > for managing Primary Node and Cluster Coordinator data are left open and > susceptible to change from any user. Also when zNodes are secured for > processor state, a “Creator Only” policy is used which allows the system to > determine the identification of the NiFi node and protect any zNodes created > with that node id using Zookeeper’s “auth” scheme. The challenge with this > scheme is that it limits the ability for other NiFi nodes in the cluster to > access that zNode if needed (since it is specifically binds that zNode to the > unique id of its creator). > > To best protect zNodes created in Zookeeper by NiFi while maximizing NiFi’s > ability to share information across the cluster I propose that we move to > using Zookeeper’s SASL authentication scheme, which will allow the use of > Kerberos principals for securing zNode with the appropriate permissions. For > maximum flexibility, these principals can be mapped appropriately in > Zookeeper, using auth-to-local rules, to ensure that nodes across the cluster > can share zNodes as needed. > > Potential Concerns/Challenges for Discussion: > > 1) For existing NiFi users how will we migrate Zookeeper entries from > the old security scheme to the new scheme? > 2) How should zNodes be reverted to open if kerberos is disabled? > 3) What will the performance impact be on the cluster once SASL scheme > is enabled (since we’d be moving from open to protected)? Would require > investigation > 4) Currently users can control authentication scheme via state > management configuration for processors yet not for clusters. Should we > still maintain the practice of allowing schemes to be configurable for > processors (with SASL being the new default)? -- This message was sent by Atlassian JIRA (v6.4.14#64029)