Skip to content

Commit 4a86337

Browse files
authored
Merge pull request #2801 from zmstone/250115-update-knonw-issues
docs: db schema merge crash will be fixed in 5.8.5
2 parents 8488ff2 + fe491bd commit 4a86337

File tree

2 files changed

+24
-20
lines changed

2 files changed

+24
-20
lines changed

en_US/changes/known-issues-5.8.md

Lines changed: 13 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -1,23 +1,13 @@
11
# Known Issues in EMQX 5.8
22

3-
## e5.8.2
3+
## e5.8.5
44

55
- **Node Crash if Linux monotonic clock steps backward (since 5.0)**
66

77
In certain virtual Linux environments, the operating system is unable to keep the clocks monotonic,
88
which may cause Erlang VM to exit with message `OS monotonic time stepped backwards!`.
99
For such environments, one may set the `+c` flag to `false` in `etc/vm.args`.
1010

11-
- **Node Cannot Start if a New Node Joined Cluster While It was Stopped (since 5.0)**
12-
13-
In a cluster of 2 or more nodes, if a new node joins the cluster while some nodes are down, the nodes which were down will fail to restart and will emit logs like below.
14-
`2024-10-03T17:13:45.063985+00:00 [error] Mnesia('[email protected]'): ** ERROR ** (core dumped to file: "/opt/emqx/[email protected]_1727_975625_63176"), ** FATAL ** Failed to merge schema: {aborted,function_clause}`
15-
16-
> **Workaround:**
17-
> Delete the `data/mnesia` directory and restart the node.
18-
19-
<!-- https://emqx.atlassian.net/browse/EMQX-12290 -->
20-
2111
- **IoTDB May Not Work Properly in Batch Mode when `batch_size > 1` (since 5.0)**
2212

2313
This issue arises because EMQX uses the IoTDB v1 API, which lacks native support for batch operations. To simulate batch functionality, an iterative approach is used; however, this method is not atomic and may lead to bugs.
@@ -28,6 +18,18 @@
2818

2919
EMQX Dashboard supports Single Sign-On based on the Security Assertion Markup Language (SAML) 2.0 standard and integrates with Okta and OneLogin as identity providers. However, the SAML-based SSO currently does not support a certificate signature verification mechanism and is incompatible with Azure Entra ID due to its complexity.
3020

21+
## e5.8.4
22+
23+
- **Node Cannot Start if a New Node Joined Cluster While It was Stopped (since 5.0, fixed in 5.8.5)**
24+
25+
In a cluster of 2 or more nodes, if a new node joins the cluster while some nodes are down, the nodes that were down will fail to restart and will emit logs like below.
26+
`2024-10-03T17:13:45.063985+00:00 [error] Mnesia('[email protected]'): ** ERROR ** (core dumped to file: "/opt/emqx/[email protected]_1727_975625_63176"), ** FATAL ** Failed to merge schema: {aborted,function_clause}`
27+
28+
> **Workaround:**
29+
> Delete the `data/mnesia` directory and restart the node.
30+
31+
<!-- https://emqx.atlassian.net/browse/EMQX-12290 -->
32+
3133
## e5.8.1
3234

3335
- **Kafka Disk Buffer Directory Name (since 5.8.0, fixed in 5.8.2)**

zh_CN/changes/known-issues-5.8.md

Lines changed: 11 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -1,19 +1,11 @@
11
# EMQX 5.8 已知问题
22

3-
## e5.8.2
3+
## e5.8.5
44

55
- **Linux 单调时钟回调导致 EMQX 节点重启 (始于 5.0)**
66

77
在某些虚拟 Linux 环境中,操作系统无法保持时钟的单调性,这可能会导致 Erlang VM 因为错误消息 `OS monotonic time stepped backwards!` 而退出。对于这类环境,可以在 `etc/vm.args` 中将 `+c` 标志设置为 `false`
88

9-
- **如果一个新节点在原节点停止时加入集群,原节点无法启动 (始于 5.0)**
10-
11-
在包含两个或更多节点的集群中,如果在某些节点停止运行时有新节点加入集群,那么这些停止的节点将无法重新启动,并会产生如下日志: `2024-10-03T17:13:45.063985+00:00 [error] Mnesia('[email protected]'): ** ERROR ** (核心转储至文件: "/opt/emqx/[email protected]_1727_975625_63176"), ** FATAL ** 合并 schema 失败: {aborted,function_clause}`
12-
13-
> **解决方法:** 删除 `data/mnesia` 目录并重新启动节点。
14-
15-
<!-- https://emqx.atlassian.net/browse/EMQX-12290 -->
16-
179
- **IoTDB 在批处理模式下(当 `batch_size > 1` 时)可能无法正常工作 (始于 5.0)**
1810

1911
出现该问题的原因是 EMQX 使用了 IoTDB v1 API,而该 API 不支持原生的批处理操作。为模拟批处理操作,系统采用了迭代方式,然而,此方法不是原子的,可能会导致出现错误。
@@ -24,6 +16,16 @@
2416

2517
EMQX Dashboard 支持基于安全断言标记语言(SAML)2.0标准的单点登录(SSO),并与 Okta 和OneLogin 作为身份提供商集成。然而,基于 SAML 的 SSO 目前不支持证书签名验证机制,并且由于其复杂性,无法与 Azure Entra ID 兼容。
2618

19+
## e5.8.4
20+
21+
- **如果一个新节点在原节点停止时加入集群,原节点无法启动 (始于 5.0,已在 5.8.5 中修复)**
22+
23+
在包含两个或更多节点的集群中,如果在某些节点停止运行时有新节点加入集群,那么这些停止的节点将无法重新启动,并会产生如下日志: `2024-10-03T17:13:45.063985+00:00 [error] Mnesia('[email protected]'): ** ERROR ** (核心转储至文件: "/opt/emqx/[email protected]_1727_975625_63176"), ** FATAL ** 合并 schema 失败: {aborted,function_clause}`
24+
25+
> **解决方法:** 删除 `data/mnesia` 目录并重新启动节点。
26+
27+
<!-- https://emqx.atlassian.net/browse/EMQX-12290 -->
28+
2729

2830
## e5.8.1
2931

0 commit comments

Comments
 (0)