curl_easy_perform: error(28): Operation timed out...

[ckcheong79]

Friends of Zimbra,

I am facing this issue almost everyday, the zimbra services are all running, but all email clients (including zimbraAdmin page) are not able to login / authenticate.

However, after zmcontrol stop and start it back to normal. But , it happens again after a day or two.

Version: Release 8.8.5_GA_1894.RHEL7_64_20171026035615 RHEL7_64 FOSS edition

Error from zimbra.log
Jan 27 14:49:46 mail saslauthd[9937]: authentication against url 'https://mail.domain.com:7073/service/admin/soap/' caused error 'curl_easy_perform: error(28): Operation timed out after 15004 milliseconds with 0 out of 0 bytes received'

Configuration from zmprov getServer mail.domain.com | grep Auth
zimbraAuthTokenNotificationInterval: 60000
zimbraLowestSupportedAuthVersion: 2
zimbraMtaAuthEnabled: TRUE
zimbraMtaAuthPort: 7073
zimbraMtaAuthTarget: TRUE
zimbraMtaBrokenSaslAuthClients: yes
zimbraMtaSaslAuthEnable: yes
zimbraMtaSmtpSaslAuthEnable: no
zimbraMtaSmtpdClientAuthRateLimit: 0
zimbraMtaSmtpdSaslAuthenticatedHeader: no
zimbraMtaTlsAuthOnly: FALSE
zimbraShareNotificationMtaAuthRequired: FALSE

Appreciated with thanks.

[phoenix]

First of all, one out-of-context line from a log file doesn't tell us much about the problem. Second, from the log line that you've posted I would have thought the problem is already identified (see the highlighted text below):

Jan 27 14:49:46 mail saslauthd[9937]: authentication against url 'https://mail.domain.com:7073/service/admin/soap/' caused error 'curl_easy_perform: error(28): Operation timed out after 15004 milliseconds with 0 out of 0 bytes received'

You've also not given any details of whether this is a new problem or it's been happening for a while nor have you given any information of the number of users on your server at the point the problem and whether they are using the Web UI or IMAP and no details of the specifications of the hardware (RAM, cpu usage, VM or bare metal etc., etc.) on which ZCS is running. You'll have to try and determine why the connection is timing out, could this be a firewall problem or a server load problem?

[ckcheong79]

Hi Bill,

This has happened since 2 weeks, but not daily.

I am running on VM environment with 6 CPUs, 8GB RAM and 600GB of SSD HDD.

There are 682 email accounts. Mostly are POP and SMTP connections, 10% mobile imap connections.

ZCS - Version: Release 8.8.5_GA_1894.RHEL7_64_20171026035615 RHEL7_64 FOSS edition

Regards n Thanks.

[phoenix]

I would suggest you have too many CPUs allocated to trhat VM and possibly not enough RAM, how many users are connected at any one time (an average or rough guess)? Is all of the ZCS directory structure on an SSD? Have you checked if there's any iowait associated with disk access on your server? Do you run any kind of monitoring software that would enable you to monitor the ZCS server performance? For vCPU resources see the following wiki article: https://wiki.zimbra.com/wiki/Performanc ... re_vSphere

[ckcheong79]

Hi Bill,

I am not sure does these two articles explain to my problem, please spare some time to read:-

https://github.com/curl/curl/issues/619

https://github.com/incarose/curl/commit ... 1bdc768cb9

Regards, and thanks.

[vargacs]

Hello,

Same issue here. Can you solve the problem?
I upgraded the ZCS from 8.7 to 8.8.6, and upgraded the ubuntu from 14.04 to 16.04. And from the next day on every day i need to restart the zmmailboxdctl because the same error. And after the restart everything is working fine for one day.
Is the bugfix can solve this issule? (zcs-patch-8.8.6_GA_HF_108839)

Thanks,
Csabi

[ckcheong79]

Hi,

I have tried using configuration of 4 CPU 8GB ram and 6 CPU 16GB ram. But still get error msg daily.

Will it because the curl has a bug? I read about this in some article. Anyone can confirm?

Thanks.

[ckcheong79]

Hi All,

I have upgraded curl version to the latest version but till no luck.

curl 7.59.0 (x86_64-redhat-linux-gnu) libcurl/7.59.0 NSS/3.28.4 zlib/1.2.7 libpsl/0.7.0 (+libicu/50.1.2) libssh2/1.8.0 nghttp2/1.21.1
Release-Date: 2018-03-14
Protocols: dict file ftp ftps gopher http https imap imaps ldap ldaps pop3 pop3s rtsp scp sftp smb smbs smtp smtps telnet tftp
Features: AsynchDNS IPv6 Largefile GSS-API Kerberos SPNEGO NTLM NTLM_WB SSL libz HTTP2 UnixSockets HTTPS-proxy PSL Metalink

Today's Error Msg from zimbra.log

Mar 19 15:07:01 mail saslauthd[28415]: authentication against url 'https://mail.hostname.com:7073/service/admin/soap/' caused error 'curl_easy_perform: error(28): Operation timed out after 15000 milliseconds with 0 out of 0 bytes received'
Mar 19 15:07:01 mail saslauthd[28415]: url 'https://mail.hostname.com:7073/service/admin/soap/' will not be used for (at least) 600 seconds
Mar 19 15:07:01 mail saslauthd[28415]: Authentication cycle re-elected url https://mail.hostname.com:7073/service/admin/soap/, giving up ...
Mar 19 15:07:01 mail saslauthd[28415]: auth_zimbra: username@domain.com.my auth failed: curl_easy_perform: error(28): Operation timed out after 15000 milliseconds with 0 out of 0 bytes received
Mar 19 15:07:01 mail saslauthd[28415]: do_auth: auth failure: [user=username@domain.com.my] [service=smtp] [realm=domain.com.my] [mech=zimbra] [reason=Unknown]
Mar 19 15:07:01 mail saslauthd[28415]: zmauth: authenticating against elected url 'https://mail.hostname.com:7073/service/admin/soap/' ...

Anyone can help? I need to perform zmcontrol stop and start everyday at least once in order to keep my server running.