Recommended Patches for Direct NFS Client (Doc ID 1495104.1)

 Patch: 32338919

Patch: 32806069

Invalid RPC frame error accompanied with I/O errors and other file open errors when using NFSv4.

ORA-600 [KSFDODM_CREATE1] while running RMAN backup with dNFS and NFSv4

12.2.0.1
18.*
19.*

ORA-7445 [KGNFSGETMNTHDL()+3418] when using Direct NFS

This bug fixes an issue in Direct NFS where 64 byte NFS file handles were not being managed correctly by Direct NFS. This problem has been observed while running NFS backups to a Data Domain NFS server but could potentially happen with any NFS server.

11.2.0.2

11.2.0.3

11.2.0.3.5 EXA BP

11.2.0.2

11.2.0.3

11.2.0.3.22 EXA BP

Improve Exadata RMAN backup performance to a ZFSSA.

Direct NFS is used to backup Exadata databases to a ZFS storage appliance. Direct NFS uses 1 MB send and receive buffers to issue 1 MB I/Os and this causes a performance degradation. The fix increases the buffer sizes to 4 MB to provide improved backup performance from an Exadata to a ZFSSA.

11.2.0.2

11.2.0.3

11.2.0.3.11 EXA BP

Intermittent slow I/O if process is idle for over 5 minutes.

Each Oracle process creates its own connection to the NFS server. If the process is idle for over 5 minutes i.e. it does not issue any I/O for over 5 minutes, the NFS server cleans up the connection and the Oracle process has to reconnect when the next I/O is issued. This bug fixes an issue with the reconnect that caused the reconnect to take longer and increase the I/O latency.

11.2.0.2

11.2.0.3

Intermittent performance degradation with Direct NFS client.

Direct NFS retries I/Os with a very short timeout and under heavy I/O load the amount of retries can go up significantly leading to wasteful retransmissions and performance degradation. This bug increases the retry interval to eliminate the performance degradation. This bug can affect all configurations and may be accompanied with a PING timeout error in certain occurences.

11.2.0.2

11.2.0.3

Intermittent performance degradation with PING timeout messages and database hang symptoms.

Direct NFS: channel id 0 path **** to filer **** PING timeout.

This bug is caused by a spin in Direct NFS when the network and I/O load is very high and I/O latencies in excess of 5 seconds are seen. This issue is only seen in multipath environments.

If the NFS storage is NetApp, the following signature is observed in the messages file on the server.

Event 'nfsd.tcp.close.idle.notify' suppressed * times. Shutting down idle connection to client (***.***.***.***) where receive side flow control has been enabled. There are *** bytes in the receive buffer. This socket is being closed from the deferred queue.

11.2.0.2

11.2.0.3

This bug is a Linux bug that affects Direct NFS configurations that do not use oranfstab. This bug affects 2.6.32 and 2.6.39 UEK Linux kernels and the bug is fixed in 2.6.32-300 and 2.6.39-200. Direct NFS users hitting this issue will see a database hang. Direct NFS is spinning in kgnfs_connect() -> skgnfs_bind() loop as the bind() call fails.

Workaround: A simple workaround for this issue is to configure oranfstab. The 'local' field should be specified in oranfstab and this will allow the bind() to succeed.

There is a small performance degradation for multiblock reads when using Direct NFS as Direct NFS fragments the reads into multiple single block reads. These reads are not fragmented if the OS NFS client is being used.

11.2.0.2

11.2.0.3

12.1.0.1

Hang observed in Direct NFS when running read intensive workloads. The hang call stack will contain one of the following functions.

skgnfs_poll_channels()<-kgnfswat()<-kgnfs_wat()<-kgodm_reap_ios()<-kgodm_io().

Multipath environments

11.2.0.2

11.2.0.3

12.1.0.1

Slow I/O performance with Direct NFS and channel down error messages in the alert log. This bug is only seen in ZFSSA environments where Direct NFS and ZFSSA are communicating over an InfiniBand network. This does not impact standard Ethernet deployments. You will also see the following messages in the database alert log, /var/log/messages and the ZFSSA logs.

alert log - Direct NFS: channel id [0] path [xxx.xxx.xxx.xxx] to filer ****** is DOWN

/var/log/messages - nfs: server ****** not responding, still trying"

ZFSSA logs - WARNING: High TCP connect timeout rate! System (port 2049) may be under a SYN flood attack.

REPLACED BY  Bug 17730745 - to track regression bug 17158897

11.2.0.2

11.2.0.3

11.2.0.4

12.1.0.1

11.2.0.2

11.2.0.3

11.2.0.4

12.1.0.1

11.2.0.2

11.2.0.3

11.2.0.4

This bug addresses an intermittent ORA-7445 seen on database startup when using Direct NFS. The error is seen with the following signature:

ORA-07445: exception encountered: core dump [kgnfs_check_for_timedout_requests()+***] [SIGSEGV]

11.2.0.3

11.2.0.4

Wrong data returned for fragmented 4 MB RMAN reads in multipath environments with high I/O latency.

If the I/O latency is over 5 seconds, read fragments may be failed over to the other path and one can potentially hit this bug where wrong results are returned. The on-disk data is still correct and there is no on-disk data corruption.

Multipath environments

11.2.0.4

12.1.0.1

Performance regression in RMAN backups and archival in 11.2.0.4 if the storage is a ZFSSA

This regression is introduced in 11.2.0.4 and the RMAN or archival throughput drops because the database is issuing synchronous writes to the ZFSSA.

Direct NFS connected to ZFSSA

11.2.0.4

Support management interface via oranfstab to issue SNMP queries to the storage.

Many storage administrators only enable SNMP over the management path. Previously, Direct NFS used to issue the SNMP checks over the data interface. With this fix, you can issue SNMP checks over the management interface.

11.2.0.2

11.2.0.3 

Performance regression seen with certain workloads when running Direct NFS on Windows.

11.2.0.4

12.1.0.2

Slow RMAN backup as copy performance from ASM to ZFS

This issue impacts the performance of RMAN backup as copy from an ASM or Exadata environment to an NFS backup location. The fix is in the RMAN layer but this is being tracked for Direct NFS since Direct NFS is the recommended backup method when backing up to NFS shares.

11.2.0.3

11.2.0.4

12.1.0.2

11.2.0.4.20 EXA BP

12.1.0.2.13 DB BP

Long Direct NFS reconnect timeout when path comes back up after failover

When a Direct NFS path goes down due to failover, Direct NFS continues using the other available paths. It checks if the other path is back up periodically but has a long timeout since connects can block. Direct NFS now supports asynchronous connects so we are reducing this timeout to reconnect quickly.

Generic

11.2.0.3

11.2.0.4

12.1.0.2

12.1.0.2.170418 EXA DB BP

Support for non-default SNMP community string values

Certain users would like to use an SNMP community string other than the default value of 'public'. This patch lets users specify their own community string for Direct NFS to use when issuing the SNMP query to the storage server.

Generic

11.2.0.3

11.2.0.4

12.1.0.1

N/A

13571798

18180390

Errors seen in alter log on Windows databases when using Direct NFS

Windows Direct NFS was not handling directory creation correctly which is essential for FRA. This bug addresses this issue.

Direct NFS on Windows

11.2.0.3

11.2.0.4

11.2.0.4.3

WIN DB BP

22515353

High I/O latency reported by OEM when using Direct NFS

The high I/O latency was being reported due to a stale timestamp resulting in incorrect latency values being computed.

Generic

11.2.0.4

12.1.0.1

12.1.0.2

N/A

22261050

Crashes seen when Direct NFS is running with NFSv4 or NFSv4.1

This fix addresses various ORA-600/ ORA-7445 errors seen when Direct NFS is running with NFSv4 or NFSv4.1.

Generic

12.1.0.2

N/A

22606521

Crashes/ hangs seen in Direct NFS on truncated read completion

This fix addresses a Direct NFS bug when handling truncated read responses or partial reads due to end-of-file.

Generic

11.2.0.4

12.1.0.2

N/A

22522515

Crashes/ hangs seen in Direct NFS when IPv6 is being used

This fix addresses issues with parsing, bind and mount when IPv6 is being used.

Generic

11.2.0.4

12.1.0.2

12.2.0.1

N/A

26203182

ORA-600 [3020] seen on standby database with Direct NFS enabled

Under heavy database I/O load, Direct NFS may end up re-transmitting I/O requests including writes to the ZFSSA. There is a slight chance of an ordering issue between a new write and a previously re-transmitted write causing a lost write. The lost write is eventually detected and reported via the ORA-600 [3020] error.

Direct NFS connected to ZFSSA

11.2.0.4

12.1.0.2

12.2.0.1

N/A

33132050

Observed a database crash due to a write failure in DBW slave after enabling NFSv4.1 on Azure ANF storage.

This fix addresses database io error when enabling dnfs with nfsv4.1

Generic

19.*

19.16

21.7

34366627

IO Hang found during stress test -- Regression of 33132050

Beneric

19.*

21.*