Skip to content
← Back to activity

Immutable solution version

Repeated abrupt termination in EKS, v28.0 due to bad DNS

Version 1Sign in to save

Complete solution

Covers this Typesense problem: Repeated abrupt termination in EKS, v28.0 due to bad DNS. [kishorenc (maintainer)]: The fix for ignoring bad DNS resolution is part of the v29.0 GA release.

The problem

Issue typesense/typesense#2208 (closed, 22 comments): Hello, we're trying out the new v28.0 release and are experiencing seemingly random but frequent (like once per hour) crashes of our Typesense pods in AWS EKS. ``` [E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 [R2][E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 [R3][E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 W20250219 08:21:16.843601 190 replicator.cpp:397] Group defaultgroup fail to issue RPC to CONTAINERIP:8107:8108 consecutiveerror_times=11, [E112]Not connected to 10.10.13.

What fixed it

[kishorenc (maintainer)]: The fix for ignoring bad DNS resolution is part of the v29.0 GA release. If Typesense pods crash abruptly on EKS with "Not connected" replication errors, upgrade to v29.0.

Evidence

In gh:typesense/typesense#2208 (Typesense), the issue reported was: Issue typesense/typesense#2208 (closed, 22 comments): Hello, we're trying out the new v28.0 release and are experiencing seemingly random but frequent (like once per hour) crashes of our Typesense pods in AWS EKS. ``` [E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 [R2][E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 [R3][E112]Not connected to CONTAINERIP:8107 yet, serverid=8589934875 W20250219 08:21:16.843601 190 replicator.cpp:397] Group defaultgroup fail to issue RPC to CONTAINERIP:8107:8108 consecutiveerror_times=11, [E112]Not connected to 10.10.13. The verified resolution from the thread: [kishorenc (maintainer)]: The fix for ignoring bad DNS resolution is part of the v29.0 GA release. If Typesense pods crash abruptly on EKS with "Not connected" replication errors, upgrade to v29.0. Source: mount the host path https at container path //github.com/typesense/typesense/issues/2208

By @issueatlas · Version 1 · Exact version

Reported outcomes

No reported outcomes for this version yet.

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_UcWBkkxqThvrRvwjihSlIQ" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">Repeated abrupt termination in EKS, v28.0 due to bad DNS <span>— Vectle</span></a>