Usually I manually issue the ticket to the accumulo process owners on all servers in the cluster. Then using a bash script, you can create a cronjob that will renew the tickets.
Run on master as accumulo user, ensuring passwordless ssh is enabled... #!/bin/bash ssh server2 "kinit -R" ssh server3 "kinit -R" ssh server3 "kinit -R" So if your tablet servers are running as user "accumulo", then I usually log in to the 'accumulo' account, perform a kinit, then start-all.sh I had a lot of issues with renewing the ticket before the expiration date, so after tweaking the principal properties, eventually I just set the ticket life for a year on that specific principal. From: [email protected] [mailto:[email protected]] On Behalf Of Hyokwon Lee Sent: Wednesday, February 26, 2014 1:40 PM To: [email protected] Subject: Re: Accumulo with Kerberos Error Hi All, I have an update to my issue. Have turning on kerberos debug logging, I have found that a new kerberos ticket is initialized after the current ticket expires. However table scans fail after the expiration of the original ticket that was issued at startup of accumulo services. In the logs I see the following errors: When trying to scan a table: ERROR: java.lang.RuntimeException: org.apache.accumulo.core.client.impl.AccumuloServerException: Error on server accumulo.test.local:9997 Numerous !METADATA -> FILE_READ -> /accumulo/tables/!0/root_tablet/xxxx Call to accumulo.test.local/127.0.0.1:8020<http://127.0.0.1:8020> failed on local exception: java.io.IOException: javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] and the log also keeps spamming: [securty.UserGroupInformation] ERROR: PrivilegedActionException as:accumulo/[email protected] (auth:KERBEROS) cause:javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] Has anyone seen this occur before? I can simply recreate this issue by letting the accumulo services start up and expire and then just watching the logs and scanning the tables. Thanks, Hokie On Tue, Feb 25, 2014 at 5:08 PM, Hyokwon Lee <[email protected]<mailto:[email protected]>> wrote: Hi Sean, The Kerberos Tickets that are being used are not renewable. Should they be? I assume even if they are after their renewable time expires I will run into the same issue? Thanks, Hokie On Tue, Feb 25, 2014 at 4:39 PM, Sean Busbey <[email protected]<mailto:[email protected]>> wrote: Hi Hokie! Are the kerberos tickets you're getting renewable? -Sean On Tue, Feb 25, 2014 at 4:35 PM, Hyokwon Lee <[email protected]<mailto:[email protected]>> wrote: I am currently running into an issue and was hoping someone may have some insight to the problem. Running Accumulo 1.4.3 on top of a Kerberos enabled Hadoop. I followed the following instructions in the README: "If you are running on top of hdfs with kerberos enabled, then you need to do some extra work. First, create an Accumulo principal kadmin.local -q "addprinc -randkey accumulo/<host.domain.name<http://host.domain.name>>" where <host.domain.name<http://host.domain.name>> is replaced by a fully qualified domain name. Export the principals to a keytab file. It is safer to create a unique keytab file for each server, but you can also glob them if you wish. kadmin.local -q "xst -k accumulo.keytab -glob accumulo*" Place this file in $ACCUMULO_HOME/conf for every host. It should be owned by the accumulo user and chmodded to 400. Add the following to the accumulo-env.sh In the accumulo-site.xml file on each node, add settings for general.kerberos.keytab and general.kerberos.principal, where the keytab setting is the absolute path to the keytab file ($ACCUMULO_HOME is valid to use) and principal is set to accumulo/_HOST@<REALM>, where REALM is set to your kerberos realm. You may use _HOST in lieu of your individual host names. <property> <name>general.kerberos.keytab</name> <value>$ACCUMULO_HOME/conf/accumulo.keytab</value> </property> <property> <name>general.kerberos.principal</name> <value>accumulo/_HOST@MYREALM</value> </property> You can then start up Accumulo as you would with the accumulo user, and it will automatically handle the kerberos keys needed to access hdfs. Please Note: You may have issues initializing Accumulo while running kerberos HDFS. You can resolve this by temporarily granting the accumulo user write access to the hdfs root directory, running init, and then revoking write permission in the root directory (be sure to maintain access to the /accumulo directory)." After doing so, got accumulo to come up and initially it states on start up that i authenticated using accumulo/[email protected]<mailto:accumulo/[email protected]>. For the next 24 hour it is happy and everything works. However after the 24 hour marker which is when the kerberos ticket expires, I start seeing the following errors on all TServers: [securty.UserGroupInformation] ERROR: PrivilegedActionException as:accumulo/[email protected] (auth:KERBEROS) cause:javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] [ipc.Client] WARN : Exception encountered while connecting to the server : javax.security.sasl.SasleEception: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] [securty.UserGroupInformation] ERROR: PrivilegedActionException as:accumulo/[email protected] (auth:KERBEROS) cause:javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] And as far as I can tell this just retries and keeps failing. I checked the accumulo.keytab file and it is a glob so it has the entries for every server that Accumulo is on. Also if I manually do a kinit -kt accumulo.keytab accumulo/[email protected]<mailto:accumulo/[email protected]> it works find and I get a valid ticket. I also made sure everything in hdfs under "/accumulo" is owned by accumulo so that doesn't seem to be the problem. Also made sure after kiniting I can access the directory path and all sub directories. So far the only thing that seems to fix my issue is if I bounce all accumulo services and it is happy again. Also until I bounce the accumulo services, I get error logs stating it cannot scan any of the tables (unable to scan metadata, root_tablet, default_tablet, etc.) Has anyone else seen this issue? Did I miss a configuration somewhere possibly? Thanks, Hokie -- __________________________________________ Hyokwon Lee [email protected]<mailto:[email protected]> -- __________________________________________ Hyokwon Lee [email protected]<mailto:[email protected]>
