openshift-docs/modules/network-observability-operator-release-notes-1-9-0-fixed-issues.adoc

// Module included in the following assemblies:
// * network_observability/release_notes_archive/network-observability-operator-release-notes-archive.adoc

:_mod-docs-content-type: REFERENCE
[id="network-observability-operator-release-notes-1-9-0-fixed-issues_{context}"]
= Network Observability Operator 1.9.0 fixed issues

[role="_abstract"]
You can review the fixed issues for the Network Observability Operator 1.9.0 release.

* Previously, when filtering by source or destination IP from the console plugin, using a Classless Inter-Domain Routing (CIDR) notation such as `10.128.0.0/24` did not work, returning results that should be filtered out. With this update, it is now possible to use a CIDR notation, with the results being filtered as expected. (link:https://issues.redhat.com/browse/NETOBSERV-2276[NETOBSERV-2276])

* Previously, network flows might have incorrectly identified the network interfaces in use, especially with a risk of mixing up `eth0` and `ens5`. This issue only occurred when the eBPF agents were configured as `Privileged`. With this update, it has been fixed partially, and almost all network interfaces are correctly identified. Refer to the known issues below for more details. (link:https://issues.redhat.com/browse/NETOBSERV-2257[NETOBSERV-2257])

* Previously, when the Operator checked for available Kubernetes APIs in order to adapt its behavior, if there was a stale API, this resulted in an error that prevented the Operator from starting normally. With this update, the Operator ignores error on unrelated APIs, logs errors on related APIs, and continues to run normally. (link:https://issues.redhat.com/browse/NETOBSERV-2240[NETOBSERV-2240])

* Previously, users could not sort flows by *Bytes* or *Packets* in the *Traffic* flows view of the Console plugin. With this update, users can sort flows by *Bytes* and *Packets*. (link:https://issues.redhat.com/browse/NETOBSERV-2239[NETOBSERV-2239])

* Previously, when configuring the `FlowCollector` resource with an IPFIX exporter, MAC addresses in the IPFIX flows were truncated to their 2 first bytes. With this update, MAC addresses are fully represented in the IPFIX flows. (link:https://issues.redhat.com/browse/NETOBSERV-2208[NETOBSERV-2208])

* Previously, some of the warnings sent from the Operator validation webhook could lack clarity on what needed to be done. With this update, some of these messages have been reviewed and amended to make them more actionable. (link:https://issues.redhat.com/browse/NETOBSERV-2178[NETOBSERV-2178])

* Previously, it was not obvious to figure out there was an issue when referencing a `LokiStack` from the `FlowCollector` resource, such as in case of typing error. With this update, the `FlowCollector` status clearly states that the referenced `LokiStack` is not found in that case. (link:https://issues.redhat.com/browse/NETOBSERV-2174[NETOBSERV-2174])

* Previously, in the console plugin *Traffic flows* view, in case of text overflow, text ellipses sometimes hid much of the text to be displayed. With this update, it displays as much text as possible. (link:https://issues.redhat.com/browse/NETOBSERV-2119[NETOBSERV-2119])

* Previously, the console plugin for network observability 1.8.1 and earlier did not work with the {product-title} 4.19 web console, making the *Network Traffic* page inaccessible. With this update, the console plugin is compatible and the *Network Traffic* page is accessible in network observability 1.9.0. (link:https://issues.redhat.com/browse/NETOBSERV-2046[NETOBSERV-2046])

* Previously, when using conversation tracking (`logTypes: Conversations` or `logTypes: All` in the `FlowCollector` resource), the *Traffic* rates metrics visible in the dashboards were flawed, wrongly showing an out-of-control increase in traffic. Now, the metrics show more accurate traffic rates. However, note that in `Conversations` and `EndedConversations` modes, these metrics are still not completely accurate as they do not include long-standing connections. This information has been added to the documentation. The default mode `logTypes: Flows` is recommended to avoid these inaccuracy. (link:https://issues.redhat.com/browse/NETOBSERV-1955[NETOBSERV-1955])