Hi!
I reply to this since there's some data in this message queue already related to my
problem:
I had 2 ipa node cluster, where the second node had been offline for some time and at some
point we had an error while trying to reboot node1 which was a Renewal Master. The issue
was that some certs had expired and after a bit of special work we got the node1 back on
track. I can spot three problems and I can't (again) figure out which one is the cause
and which one I should repair first.
Now I got assigned the case to get the node2 back on track also. It had some certificates
expired (obviously) so I did a small time jump and some of the certs were renewed. However
not all of them were upgraded. "getcert list" reports 3 certs "CA
Unreachable", other 3 certs seem fine.
--
getcert list |grep -A 10 "CA_UNREACH"
status: CA_UNREACHABLE
ca-error: Internal error
stuck: no
key pair storage:
type=NSSDB,location='/etc/pki/pki-tomcat/alias',nickname='ocspSigningCert
cert-pki-ca',token='NSS Certificate DB',pin set
certificate:
type=NSSDB,location='/etc/pki/pki-tomcat/alias',nickname='ocspSigningCert
cert-pki-ca',token='NSS Certificate DB'
CA: dogtag-ipa-ca-renew-agent
issuer: CN=Certificate Authority,O=<<REALM>>
subject: CN=OCSP Subsystem,O=<<REALM>>
expires: 2018-03-21 09:42:04 UTC
key usage: digitalSignature,nonRepudiation,keyCertSign,cRLSign
eku: id-kp-OCSPSigning
--
status: CA_UNREACHABLE
ca-error: Internal error
stuck: no
key pair storage:
type=NSSDB,location='/etc/httpd/alias',nickname='ipaCert',token='NSS
Certificate DB',pinfile='/etc/httpd/alias/pwdfile.txt'
certificate:
type=NSSDB,location='/etc/httpd/alias',nickname='ipaCert',token='NSS
Certificate DB'
CA: dogtag-ipa-ca-renew-agent
issuer: CN=Certificate Authority,O=<<REALM>>
subject: CN=IPA RA,O=<<REALM>>
expires: 2018-03-21 09:42:29 UTC
key usage: digitalSignature,nonRepudiation,keyEncipherment,dataEncipherment
eku: id-kp-serverAuth,id-kp-clientAuth
--
status: CA_UNREACHABLE
ca-error: Error 7 connecting to
http://<<ipa2.fqdn>>:8080/ca/ee/ca/profileSubmit: Couldn't connect to
server.
stuck: no
key pair storage:
type=NSSDB,location='/etc/pki/pki-tomcat/alias',nickname='Server-Cert
cert-pki-ca',token='NSS Certificate DB',pin set
certificate:
type=NSSDB,location='/etc/pki/pki-tomcat/alias',nickname='Server-Cert
cert-pki-ca',token='NSS Certificate DB'
CA: dogtag-ipa-renew-agent
issuer: CN=Certificate Authority,O=<<REALM>>
subject: CN=<<ipa2.fqdn>>,O=<<REALM>>
expires: 2018-06-27 07:01:38 UTC
key usage: digitalSignature,nonRepudiation,keyEncipherment,dataEncipherment
eku: id-kp-serverAuth,id-kp-clientAuth,id-kp-emailProtection
--
Seems like "Server-Cert cert-pki-ca" is trying to renew on itself (node2) but
shouldn't node1 be the renewal master? Restarting httpd, certmonger and pki-tomcat
don't seem to help, time traveling helped on other certs but not on these.
Directory service seems to work if I start it manually but ipa-server-upgrade fails on
directory server not starting with "No ports specified" so something wrong with
it or is it the certificates?
--
ipa-server-upgrade
Upgrading IPA:. Estimated time: 1 minute 30 seconds
[1/10]: stopping directory server
[2/10]: saving configuration
[3/10]: disabling listeners
[4/10]: enabling DS global lock
[5/10]: starting directory server
--
<<ipa2.fqdn>> ns-slapd[24503]: [04/Jul/2018:13:43:48.829927675 +0300] - EMERG
- main - Fatal Error---No ports specified. Exiting now.
--
Also certmonger has issues:
--
dogtag-ipa-ca-renew-agent-submit[1892]: Traceback (most recent call last):
File "/usr/libexec/certmonger/dogtag-ipa-ca-renew-agent-submit", line 541, in
<module>
sys.exit(main())
File "/usr/libexec/certmonger/dogtag-ipa-ca-renew-agent-submit", line 515, in
main
kinit_keytab(principal, paths.KRB5_KEYTAB, ccache_filename)
File "/usr/lib/python2.7/site-packages/ipalib/install/kinit.py", line 43, in
kinit_keytab
cred = gssapi.Credentials(name=name, store=store, usage='initiate')
File "/usr/lib64/python2.7/site-packages/gssapi/creds.py", line 64, in __new__
store=store)
File "/usr/lib64/python2.7/site-packages/gssapi/creds.py", line 148, in
acquire
usage)
File "ext_cred_store.pyx", line 182, in
gssapi.raw.ext_cred_store.acquire_cred_from (gssapi/raw/ext_cred_store.c:1732)
GSSError: Major (851968): Unspecified GSS failure. Minor code may provide more
information, Minor (2529639068): Cannot contact any KDC for realm
'<<REALM>>'
--
but KDCs should be able to be resolved even from ipa node2
--
nslookup -type=srv _kerberos._tcp.<<REALM>>
Server: <<ipa1.ip>>
Address: <<ipa1.ip>>#53
_kerberos._tcp.<<REALM>> service = 0 100 88 <<ipa1.fqdn>>.
_kerberos._tcp.<<REALM>> service = 0 100 88 <<ipa2.fqdn>>.
--
For testing purposes I turned off firewall on ipa node1
Eemeli
-----Original Message-----
From: Rob Crittenden [mailto:rcritten@redhat.com]
Sent: torstai 28. kesäkuuta 2018 16.05
To: Jokinen Eemeli <Eemeli.Jokinen(a)cinia.fi>; FreeIPA users list
<freeipa-users(a)lists.fedorahosted.org>; Florence Blanc-Renaud
<flo(a)redhat.com>
Subject: Re: [Freeipa-users] Re: Problems after IPA upgrade: ipa-server-upgrade
doesn't complete, pki-tomcatd won't start
Jokinen Eemeli wrote:
Hi!
No I haven't since my guide line didn't tell me to.
I tried to set the date back, restart certmonger and then I did "ipactl
restart" and then it got 2 certs renewed! One of the remaining two certificates was
on "CA_UNREACHABLE" state, so I ran another certmonger restart and it did get
updated. The last one didn't seem to go anywhere so I resubmitted the cert request and
then that one also got renewed. I time jumped back to today and did another ipactl restart
and All this mess got started with failed "ipa-server-upgrade" so I ran it
afterwards and it completed successfully with no errors. That also increased the number of
certmonger tracked certificates to 9 from 8 so I believe that one is fixed too.
There are layers of dependencies on the certs so sometimes multiple rounds of renewal are
needed to sort things out. This normally happens gracefully as expiration approaches but
in some cases that we haven't been able to identify this doesn't happen.
Thank you a lot! It's a bit complicated mess to understand every aspect of it (for
example I was trying to hunt missing certificate that certmonger didn't track even
though it wasn't the issue but the outcome of failed server upgrade) but after this I
believe very that I understand it a way better!
Cool, glad you are back up and running.
Note that the cert issues weren't caused by the upgrade, the upgrade just made it more
apparent.
In order to be sure the upgrade is complete you should run: # ipa-server-upgrade
The upgrade will also check all of the certs tracked by certmonger and ensure they are set
up correctly.
rob
Eemeli
-----Original Message-----
From: Rob Crittenden [mailto:rcritten@redhat.com]
Sent: keskiviikko 27. kesäkuuta 2018 16.26
To: FreeIPA users list <freeipa-users(a)lists.fedorahosted.org>;
Florence Blanc-Renaud <flo(a)redhat.com>
Cc: Jokinen Eemeli <Eemeli.Jokinen(a)cinia.fi>
Subject: Re: [Freeipa-users] Re: Problems after IPA upgrade:
ipa-server-upgrade doesn't complete, pki-tomcatd won't start
Hard to know without seeing the list of certs.
Are you restarting dogtag, Apache and 389-ds when setting the date back?
That is necessary as well.
rob