Sidebar

7 已知问题

全局事件关联

如果第一次和第二次事件之间的时间间隔非常短,即半秒或更短,则事件可能无法正确关联。

IPMI 检查

在 Debian 9(stretch)之前和 Ubuntu16.04(xenial)之前使用 OpenIPMI 库,IPMI 检查可能无法正常工作。若要解决此问题,需要重新编译 OpenIPMI 库并启用 OpenSSL,详见ZBX-6139

SSH 检查

一些 Linux 发行版本如 Debian、Ubuntu,如果使用了安装包安装了 libssh2 类库,则系统将不支持使用密码加密私钥,详见 ZBX-4850 获得更多信息。

ODBC 检查

由于 upstream bug,如果 Zabbix server 或 proxy 使用 MySQL 作为其数据库,MySQL ODBC 库可能无法使用。有关更多信息和可用的解决办法,详见 ZBX-7665

由于 Microsoft 的 问题。从 Microsoft SQL Server 查询的 XML 数据可能会被截断为 2033 个字符。

HTTPS 检查

在使用 https 协议的 Web 场景和 HTTP agent 监控项,如果目标服务器配置了禁止 TLS v1.0 或更低版本的协议,Zabbix agent 检查 net.tcp.service[https...]net.tcp.service.perf[https...] 可能会失败。有关更多信息和可用的解决方法,详见ZBX-9879

Web 监控和 HTTP agent

当 "SSL verify peer" 在 Web 场景或 HTTP agent 启用时,由于upstream bug,Zabbix server 可能在 CentOS6、CentOS7和其他相关 Linux 发行版本上发生内存泄露。有关更多信息和可用的解决方法,详见ZBX-10486

简单检查

由于早于 v3.10 和 2.1.2 版本的 fping 存在一个 BUG,即它错误地处理重复的回放数据包。这可能会使监控项 icmppingicmppinglossicmppingsec 导致一些意外的结果。建议使用最新版本的fping。详见 ZBX-11726 获得更多信息。

SNMP 检查

如果使用 OpenBSD 操作系统,并在 Zabbix server 的配置文件中设置了 SourceIP 参数,则在 5.7.3 版本的 Net-SNMP 库中的一个 Use-After-Free(UAF)漏洞可能导致Zabbix server 崩溃。 作为解决方法,请不要设置 SourceIP 参数。 同样的问题也适用于Linux,但它不会导致 Zabbix server 停止工作。应用与 OpenBSD 上 net-snmp 软件包的局部补丁,将会随 OpenBSD 6.3 版本一起发布。

PHP 7.0 的兼容性问题

已经观察到,使用 PHP 7.0 导入具有 Web 监控触发器的模板,可能会因触发器表达式中的 Web 监控项的双引号错误而导入失败。 但将 PHP 升级到 7.1 时,问题就随之消失了。

图表

切换到夏令时(Daylight Saving Time,DST)会导致显示 X 轴标签错误(如日期重复,日期缺失等)。

日志文件监控

当文件系统空间为 100% 已满时,如果日志文件仍然在被追加,那么 log[]logrt[] 监控想会反复从头重新读取日志文件。详见 ZBX-10884 获得更多信息。

MySQL 的慢查询

如果监控项的值不存在,那么 Zabbix server 将会生成慢查询(关于 SELECT)。这是由于 MySQL 5.6/5.7 版本中一个已知的问题造成的。解决此问题的办法是在 MySQL 中禁用 index_condition_pushdown 优化器。详见ZBX-10652

API

如果使用 history.get 方法,则 output 参数将无法正常工作。

API login

当使用带有 user.login 方法 的自定义脚本时,则可以创建大量开放式用户会话,而无需遵循 user.logout

IPv6 address issue in SNMPv3 traps

由于net-snmp错误,在SNMP陷阱中使用SNMPv3时可能无法正确显示IPv6地址。有关更多详细信息和可能的解决方法,参见 ZBX-14541.

4.0.0的已知问题

  • 无法使用相应的URL参数访问仪表板的全屏模式:/zabbix.php?action=dashboard.view&fullscreen=1。在4.0.1中已经被修复。
  • 监控 - > 概述(触发器)中过滤触发器时,会考虑触发器严重性而不是问题(problem)严重性。\

Flipping frontend locales

It has been observed that frontend locales may flip without apparent logic, i. e. some pages (or parts of pages) are displayed in one language while other pages (or parts of pages) in a different language. Typically the problem may appear when there are several users, some of whom use one locale, while others use another.

A known workaround to this is to disable multithreading in PHP and Apache.

The problem is related to how setting the locale works in PHP: locale information is maintained per process, not per thread. So in a multi-thread environment, when there are several projects run by same Apache process, it is possible that the locale gets changed in another thread and that changes how data can be processed in the Zabbix thread.

For more information, please see related problem reports:

  • ZBX-10911 (Problem with flipping frontend locales)
  • ZBX-16297 (Problem with number processing in graphs using the bcdiv function of BC Math functions)

Compatibility issue with PHP 7.0

It has been observed that with PHP 7.0 importing a template with web monitoring triggers may fail due to incorrectly added double quotes to the web monitoring items in the trigger expressions. The issue goes away when upgrading PHP to 7.1.

PHP 7.3 opcache configuration

If "opcache" is enabled in the PHP 7.3 configuration, Zabbix frontend may show a blank screen when loaded for the first time. This is a registered PHP bug. To work around this, please set the "opcache.optimization_level" parameter to 0x7FFFBFDF in the PHP configuration (php.ini file).

Graphs

Changes to Daylight Saving Time (DST) result in irregularities when displaying X axis labels (date duplication, date missing, etc).

Log file monitoring

log[] and logrt[] items repeatedly reread log file from the beginning if file system is 100% full and the log file is being appended (see ZBX-10884 for more information).

Slow MySQL queries

Zabbix server generates slow select queries in case of non-existing values for items. This is caused by a known issue in MySQL 5.6/5.7 versions. A workaround to this is disabling the index_condition_pushdown optimizer in MySQL. For an extended discussion, see ZBX-10652.

MariaDB row size error

Prior to MariaDB 10.2.26, 10.3.17, and 10.4.7, MariaDB didn't properly calculate the row sizes while executing DDL, so "unsafe" tables could be created, even with innodb_strict_mode=ON set. This was fixed by MDEV-19292. As a side effect, tables that could be created in previous versions may get rejected after the latest releases.

Suggested database changes for existing 4.0 installations:

shell> mysql -uroot -p<password>
       mysql> alter table `host_inventory` modify `name` varchar(128) default '' not null, modify `alias` varchar(128) default '' not null, modify `os` varchar(128) default '' not null, modify `os_short`   varchar(128) default '' not null;

Suggested database changes for new 4.0 installations:

shell> mysql -uroot -p<password>
       mysql> set global innodb_strict_mode='OFF';

Perform database creation as usual and fix schema after:

shell> mysql -uroot -p<password>
       mysql> alter table `host_inventory` modify `name` varchar(128) default '' not null, modify `alias` varchar(128) default '' not null, modify `os` varchar(128) default '' not null, modify `os_short` varchar(128) default '' not null;
       mysql> set global innodb_strict_mode='ON';

API

The output parameter does not work properly with the history.get method (fixed in 4.0.11).

API login

A large number of open user sessions can be created when using custom scripts with the user.login method without a following user.logout.

IPv6 address issue in SNMPv3 traps

Due to a net-snmp bug, IPv6 address may not be correctly displayed when using SNMPv3 in SNMP traps. For more details and a possible workaround, see ZBX-14541.

IE11 issue with map resizing in dashboard widgets

Maps, in Internet Explorer 11, are cut off on the right side if the map content is larger than the dashboard widget area (instead of being resized proportionately). This is intentional because of an IE11-related issue with proper resizing of SVG images.

Zabbix agent checks on Windows

Non-existing DNS entries in a Server parameter of Zabbix agent configuration file (zabbix_agentd.conf) may increase Zabbix agent response time on Windows. This happens because Windows DNS caching daemon doesn't cache negative responses for IPv4 addresses. However, for IPv6 addresses negative responses are cached, so a possible workaround to this is disabling IPv4 on the host.

MySQL custom error codes

If Zabbix is used with MySQL installation on Azure, an unclear error message [9002] Some errors occurred may appear in Zabbix logs. This generic error text is sent to Zabbix server or proxy by the database. To get more information about the cause of the error, check Azure logs.

Known issues in 4.0.0

  • Fullscreen mode for the dashboards cannot be accessed with the respective URL parameter: /zabbix.php?action=dashboard.view&fullscreen=1. Fixed in 4.0.1.
  • When filtering triggers in MonitoringOverview (Triggers), trigger severity instead of problem severity is taken into account.

Known issues in 4.0.0, 4.0.1

  • Upgrade to 4.0.0 and 4.0.1 from Zabbix 3.0 does not replace {TRIGGER.NAME} macros to {EVENT.NAME} macros like it should. Fixed for 4.0.2.

Known issues in 4.0.0 - 4.0.2

  • If the Check now request is performed from the item list for items that do not support it (template items, active agent items, etc.), a 'Request sent successfully' message is displayed even though the functionality does not work for these items. Fixed in 4.0.3, by displaying a corresponding error message.

Known issues in 4.0.2

  • Parent host group selection does not work in the following locations:
    • MonitoringProblems filter
    • MonitoringDashboard widgets with host group selection:
    • ReportsTriggers top 100 filter

Known issues in 4.0.0 - 4.0.3

Known issues in 4.0.3

  • If an item is deleted that has a low-level discovery created trigger, the trigger is not deleted. This leads to "orphaned" problems (with no item/host) being displayed in the problem view. Similarly, if an item is deleted that has low-level discovery created dependent item, the dependent item is not deleted. Fixed in 4.0.4.

Known issues starting with 4.0.3

  • Failed login attempt message has been changed to display only the first 39 characters of a stored IP address as that's the character limit in the database field. That means that IPv6 IP addresses longer than 39 characters will be shown incompletely.

Known issues in 4.0.0 - 4.0.5

  • net.dns[] Zabbix agent item does not support IPv6 addresses in its first parameter. Fixed in 4.0.6.

Known issues in 4.0.6, 4.0.7

  • IPMI pollers are unable to process more than 60 items per minute. Fixed in 4.0.8.
  • You may encounter the issue of 100% CPU usage by IPMI pollers if using OpenIPMI version 2.0.26 and above. Fixed in 4.0.8.

Known issues in 4.0.8

  • This version requires write permissions to the assets directory for the web server user. This requirement, however, would break those installations of the frontend where the web server has no write permissions to the document root (see ZBX-16180 for details). As a result, this requirement is rolled back in 4.0.9.

Known issues in 4.0.0 - 4.0.15

  • High memory usage has been observed during process startup with SQLite 3.7.17 on Centos/RHEL 7. The startup process has been improved in 4.0.16 to avoid similar issues with other databases as well. See ZBX-9084 for more details.