Humio Server 1.6.11 LTS (2020-01-06)

Version?Type?Release Date?Availability?End of Support

Security

Updates

Upgrades

From?

Config.

Changes?
1.6.11LTS2020-01-06

Cloud

2020-11-30No1.5.19No

Hide file hashes

Show file hashes

These notes include entries from the following previous releases: 1.6.8, 1.6.9, 1.6.10

Handle Large Global Snapshot File

Breaking Changes

The following items create a breaking change in the behavior, response or operation of this release.

  • Summary

    • LDAP: It is now possible to specify an attribute within the LDAP record to use for the username rather than the default (an email address). This is only the case when using ldap-search method by specifying the LDAP_USERNAME_ATTRIBUTE in the environment. Group names when using LDAP have historically been the distinguished name (DN) for that group, it is now possible to specify and attribute in the group record for the name by setting LDAP_GROUPNAME_ATTRIBUTE. These changes necessitated a breaking change in the ldap-search code path in cases where users of Humio authenticate with a username (e.g. user) rather than an email address (e.g. user@example.com). To elicit the same behavior as previous versions of Humio simply specify the LDAP_SEARCH_DOMAIN_NAME which in the past would default to the value of LDAP_DOMAIN_NAME but no longer does.

Fixed in this release

  • Summary

    • New background job: Find segments that are too small compared to the desired sizes (from current config) and merge them into larger files. For COMPRESSION_TYPE=high this will recompress the inputs while combining them. This job runs by default.

    • Improved memory usage from having large global.

    • Require setting LDAP_SEARCH_DOMAIN_NAME explicitly when using ldap-search authentication method.

    • Segment merge could leave out some parts when merging, leading to segments not on average becoming a large as is desired.

    • Add LDAP_USERNAME_ATTRIBUTE and LDAP_GROUPNAME_ATTRIBUTE configuration settings to enable more control over names carried from LDAP into Humio.

    • Query sessions were not properly cleaned up after becoming unused. This lead to a leak causing high amount of chatter between nodes.

    • Handle large global snapshot files (larger than 2 G).

    • Detect when event ingested are more than MAX_HOURS_SEGMENT_OPEN (24h by default) old and add the tag humioBackfill to them in that case to keep "old" events from getting mixed with current "live" events.

    • Support for "sticky autosharding" and listing of current autosharding settings for all datasources in a repository.

    • Username/email is treated case-insensitive in Humio. This is more expected behavior of usernames as emails addresses are often used. In some rare occasions duplicate accounts might have been created with difference in casing and this change can trigger the otherwise dormant account to be chosen when logging in the next time. If this happens, use the administrations page to delete the unwanted user account and let the user log in again.