Bug 1334287

Summary:	Under high read load, sometimes the message "XDR decoding failed" appears in the logs and read fails
Product:	[Community] GlusterFS	Reporter:	Xavi Hernandez <jahernan>
Component:	rpc	Assignee:	Xavi Hernandez <jahernan>
Status:	CLOSED CURRENTRELEASE	QA Contact:
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	3.8.0	CC:	bugs, florian.philippon, rgowdapp
Target Milestone:	---	Keywords:	Reopened, Triaged
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	glusterfs-3.8rc2	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:	1334285	Environment:
Last Closed:	2016-06-16 14:05:31 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1334285, 1334985
Bug Blocks:

Description Xavi Hernandez 2016-05-09 10:42:33 UTC

+++ This bug was initially created as a clone of Bug #1334285 +++

+++ This bug was initially created as a clone of Bug #1331502 +++

Description of problem:

Under some circumstances, the following errors appear in the logs:

[2016-04-28 16:48:58.710106] E [MSGID: 114030] [client-rpc-fops.c:3022:client3_3_readv_cbk] 0-test-client-2: XDR decoding failed [Invalid argument]
[2016-04-28 16:48:58.710223] W [MSGID: 114031] [client-rpc-fops.c:3050:client3_3_readv_cbk] 0-test-client-2: remote operation failed [Invalid argument]

This causes the read to fail.

Apparently there are no problems in the bricks.

Version-Release number of selected component (if applicable): 3.7.11


How reproducible:

It appears randomly but quite consistently with the following steps.

Steps to Reproduce:
1. create a distributed-disperse 4*(4+2) volume
2. export it with ganesha using gfapi
3. mount it from another machine using NFSv4
4. iozone -t2 -s10g -r1024k -i0 -w -F <mount>/iozone{1..2}.dat
5. echo 3 >/proc/sys/vm/drop_caches
6. iozone -t2 -s10g -r1024k -i1 -w -F <mount>/iozone{1..2}.dat

Generally the problem appears soon after the step 6 begins.

Actual results:

The iozone read test fails.

Expected results:

The iozone test should complete successfully.

Additional info:

This seems caused by an incorrect handling of incoming TCP data. This info is at the moment that the error happens:

Breakpoint 1, client3_3_readv_cbk (req=0x7f540e64106c, iov=0x7f540e6410ac, count=<optimized out>, myframe=0x7f54259a4d54) at client-rpc-fops.c:3021
3021                    gf_msg (this->name, GF_LOG_ERROR, EINVAL,
(gdb) print *iov
$1 = {iov_base = 0x7f53e994e018, iov_len = 116}
(gdb) x/116xb 0x7f53e994e018
0x7f53e994e018: 0x00    0x00    0x80    0x00    0x00    0x00    0x00    0x00
0x7f53e994e020: 0xa8    0xbf    0xa3    0xe0    0x5f    0x48    0x4c    0x1e
0x7f53e994e028: 0x80    0xa3    0x8a    0xd8    0x9d    0xa1    0x1c    0x75
0x7f53e994e030: 0x80    0xa3    0x8a    0xd8    0x9d    0xa1    0x1c    0x75
0x7f53e994e038: 0x00    0x00    0x00    0x00    0x00    0x00    0x08    0x00
0x7f53e994e040: 0x00    0x00    0x81    0xa0    0x00    0x00    0x00    0x01
0x7f53e994e048: 0x00    0x00    0x00    0x00    0x00    0x00    0x00    0x00
0x7f53e994e050: 0x00    0x00    0x00    0x00    0x00    0x00    0x00    0x00
0x7f53e994e058: 0x00    0x00    0x00    0x00    0xa0    0x00    0x00    0x00
0x7f53e994e060: 0x00    0x00    0x10    0x00    0x00    0x00    0x00    0x00
0x7f53e994e068: 0x00    0x50    0x00    0x00    0x57    0x22    0x04    0x1f
0x7f53e994e070: 0x25    0x38    0x92    0x91    0x57    0x22    0x04    0xb3
0x7f53e994e078: 0x03    0x53    0x1b    0x13    0x57    0x22    0x04    0xb3
0x7f53e994e080: 0x06    0xf5    0xe1    0x99    0x00    0x00    0x80    0x00
0x7f53e994e088: 0x00    0x00    0x00    0x26
(gdb) x/154xb 0x7f53e994e018
0x7f53e994e018: 0x00    0x00    0x80    0x00    0x00    0x00    0x00    0x00
0x7f53e994e020: 0xa8    0xbf    0xa3    0xe0    0x5f    0x48    0x4c    0x1e
0x7f53e994e028: 0x80    0xa3    0x8a    0xd8    0x9d    0xa1    0x1c    0x75
0x7f53e994e030: 0x80    0xa3    0x8a    0xd8    0x9d    0xa1    0x1c    0x75
0x7f53e994e038: 0x00    0x00    0x00    0x00    0x00    0x00    0x08    0x00
0x7f53e994e040: 0x00    0x00    0x81    0xa0    0x00    0x00    0x00    0x01
0x7f53e994e048: 0x00    0x00    0x00    0x00    0x00    0x00    0x00    0x00
0x7f53e994e050: 0x00    0x00    0x00    0x00    0x00    0x00    0x00    0x00
0x7f53e994e058: 0x00    0x00    0x00    0x00    0xa0    0x00    0x00    0x00
0x7f53e994e060: 0x00    0x00    0x10    0x00    0x00    0x00    0x00    0x00
0x7f53e994e068: 0x00    0x50    0x00    0x00    0x57    0x22    0x04    0x1f
0x7f53e994e070: 0x25    0x38    0x92    0x91    0x57    0x22    0x04    0xb3
0x7f53e994e078: 0x03    0x53    0x1b    0x13    0x57    0x22    0x04    0xb3
0x7f53e994e080: 0x06    0xf5    0xe1    0x99    0x00    0x00    0x80    0x00
0x7f53e994e088: 0x00    0x00    0x00    0x26    0x00    0x00    0x00    0x01
0x7f53e994e090: 0x00    0x00    0x00    0x17    0x00    0x00    0x00    0x02
0x7f53e994e098: 0x67    0x6c    0x75    0x73    0x74    0x65    0x72    0x66
0x7f53e994e0a0: 0x73    0x2e    0x69    0x6e    0x6f    0x64    0x65    0x6c
0x7f53e994e0a8: 0x6b    0x2d    0x63    0x6f    0x75    0x6e    0x74    0x00
0x7f53e994e0b0: 0x31    0x00
(gdb) bt
#0  client3_3_readv_cbk (req=0x7f540e64106c, iov=0x7f540e6410ac, count=<optimized out>, myframe=0x7f54259a4d54) at client-rpc-fops.c:3021
#1  0x00007f542a677ab0 in rpc_clnt_handle_reply (clnt=clnt@entry=0x7f54101cdef0, pollin=pollin@entry=0x7f54000091f0) at rpc-clnt.c:764
#2  0x00007f542a677d6f in rpc_clnt_notify (trans=<optimized out>, mydata=0x7f54101cdf20, event=<optimized out>, data=0x7f54000091f0) at rpc-clnt.c:925
#3  0x00007f542a673853 in rpc_transport_notify (this=this@entry=0x7f54101ddb70, event=event@entry=RPC_TRANSPORT_MSG_RECEIVED, data=data@entry=0x7f54000091f0) at rpc-transport.c:546
#4  0x00007f541d881666 in socket_event_poll_in (this=this@entry=0x7f54101ddb70) at socket.c:2237
#5  0x00007f541d8842c4 in socket_event_handler (fd=fd@entry=30, idx=idx@entry=20, data=0x7f54101ddb70, poll_in=1, poll_out=0, poll_err=0) at socket.c:2350
#6  0x00007f542a90aa4a in event_dispatch_epoll_handler (event=0x7f540effc540, event_pool=0xd3d9a0) at event-epoll.c:575
#7  event_dispatch_epoll_worker (data=0x7f54100d0890) at event-epoll.c:678
#8  0x00007f542d0fbdc5 in start_thread () from /lib64/libpthread.so.0
#9  0x00007f542c7bc28d in clone () from /lib64/libc.so.6

The fragment to decode has 116 bytes, but it should have 154 bytes (otherwise the xdata encoded is cut). I have verified that the remaining 38 bytes are stored in the beggining of the second iovec entry (iov[1]).

A network capture of this exact packet reveals the following data:

Frame 334067: 53754 bytes on wire (430032 bits), 53754 bytes captured (430032 bits)
    Encapsulation type: Ethernet (1)
    Arrival Time: Apr 28, 2016 18:00:47.706023000 Hora de verano romance
    [Time shift for this packet: 0.000000000 seconds]
    Epoch Time: 1461859247.706023000 seconds
    [Time delta from previous captured frame: 0.000130000 seconds]
    [Time delta from previous displayed frame: 0.002272000 seconds]
    [Time since reference or first frame: 5.865347000 seconds]
    Frame Number: 334067
    Frame Length: 53754 bytes (430032 bits)
    Capture Length: 53754 bytes (430032 bits)
    [Frame is marked: False]
    [Frame is ignored: False]
    [Protocols in frame: eth:ethertype:ip:tcp:rpc:glusterfs:data:rpc]
Ethernet II, Src: 00:1b:21:91:aa:30, Dst: 00:1b:21:91:aa:17
    Destination: 00:1b:21:91:aa:17
        Address: 00:1b:21:91:aa:17
        .... ..0. .... .... .... .... = LG bit: Globally unique address (factory default)
        .... ...0 .... .... .... .... = IG bit: Individual address (unicast)
    Source: 00:1b:21:91:aa:30
        Address: 00:1b:21:91:aa:30
        .... ..0. .... .... .... .... = LG bit: Globally unique address (factory default)
        .... ...0 .... .... .... .... = IG bit: Individual address (unicast)
    Type: IPv4 (0x0800)
Internet Protocol Version 4, Src: 193.109.174.232, Dst: 193.109.174.230
    0100 .... = Version: 4
    .... 0101 = Header Length: 20 bytes
    Differentiated Services Field: 0x00 (DSCP: CS0, ECN: Not-ECT)
        0000 00.. = Differentiated Services Codepoint: Default (0)
        .... ..00 = Explicit Congestion Notification: Not ECN-Capable Transport (0)
    Total Length: 53740
    Identification: 0x517a (20858)
    Flags: 0x02 (Don't Fragment)
        0... .... = Reserved bit: Not set
        .1.. .... = Don't fragment: Set
        ..0. .... = More fragments: Not set
    Fragment offset: 0
    Time to live: 64
    Protocol: TCP (6)
    Header checksum: 0x36e7 [validation disabled]
        [Good: False]
        [Bad: False]
    Source: 193.109.174.232
    Destination: 193.109.174.230
    [Source GeoIP: Unknown]
    [Destination GeoIP: Unknown]
Transmission Control Protocol, Src Port: 49152 (49152), Dst Port: 65205 (65205), Seq: 383891321, Ack: 1959945, Len: 53688
    Source Port: 49152
    Destination Port: 65205
    [Stream index: 11]
    [TCP Segment Len: 53688]
    Sequence number: 383891321    (relative sequence number)
    [Next sequence number: 383945009    (relative sequence number)]
    Acknowledgment number: 1959945    (relative ack number)
    Header Length: 32 bytes
    Flags: 0x010 (ACK)
        000. .... .... = Reserved: Not set
        ...0 .... .... = Nonce: Not set
        .... 0... .... = Congestion Window Reduced (CWR): Not set
        .... .0.. .... = ECN-Echo: Not set
        .... ..0. .... = Urgent: Not set
        .... ...1 .... = Acknowledgment: Set
        .... .... 0... = Push: Not set
        .... .... .0.. = Reset: Not set
        .... .... ..0. = Syn: Not set
        .... .... ...0 = Fin: Not set
        [TCP Flags: *******A****]
    Window size value: 17518
    [Calculated window size: 17518]
    [Window size scaling factor: -1 (unknown)]
    Checksum: 0xb289 [validation disabled]
        [Good Checksum: False]
        [Bad Checksum: False]
    Urgent pointer: 0
    Options: (12 bytes), No-Operation (NOP), No-Operation (NOP), Timestamps
        No-Operation (NOP)
            Type: 1
                0... .... = Copy on fragmentation: No
                .00. .... = Class: Control (0)
                ...0 0001 = Number: No-Operation (NOP) (1)
        No-Operation (NOP)
            Type: 1
                0... .... = Copy on fragmentation: No
                .00. .... = Class: Control (0)
                ...0 0001 = Number: No-Operation (NOP) (1)
        Timestamps: TSval 192225006, TSecr 192258004
            Kind: Time Stamp Option (8)
            Length: 10
            Timestamp value: 192225006
            Timestamp echo reply: 192258004
    [SEQ/ACK analysis]
        [Bytes in flight: 53688]
    TCP segment data (32836 bytes)
    TCP segment data (20852 bytes)
[2 Reassembled TCP Segments (32952 bytes): #334064(116), #334067(32836)]
    [Frame: 334064, payload: 0-115 (116 bytes)]
    [Frame: 334067, payload: 116-32951 (32836 bytes)]
    [Segment count: 2]
    [Reassembled TCP length: 32952]
    [Reassembled TCP Data: 800080b40000896800000001000000000000000000000000...]
Remote Procedure Call, Type:Reply XID:0x00008968
    Fragment header: Last fragment, 32948 bytes
        1... .... .... .... .... .... .... .... = Last Fragment: Yes
        .000 0000 0000 0000 1000 0000 1011 0100 = Fragment Length: 32948
    XID: 0x00008968 (35176)
    Message Type: Reply (1)
    [Program: GlusterFS (1298437)]
    [Program Version: 330]
    [Procedure: READ (12)]
    Reply State: accepted (0)
    [This is a reply to a request in frame 333872]
    [Time from request: 0.002272000 seconds]
    Verifier
        Flavor: AUTH_NULL (0)
        Length: 0
    Accept State: RPC executed successfully (0)
GlusterFS
    [Program Version: 330]
    [GlusterFS: READ (12)]
    Return value: 32768
    Errno: 0 (No error)
    IATT
        GFID: c38e35f0-31a1-4501-8a21-064b084c59df
        Inode: 9953243570797042143
        Device: 0x0000000000000800
        Mode: 0100640, S_IRUSR, S_IWUSR, S_IRGRP
            .... .... .... .... .... 0... .... .... = S_ISUID: Not set
            .... .... .... .... .... .0.. .... .... = S_ISGID: Not set
            .... .... .... .... .... ..0. .... .... = S_ISVTX: Not set
            .... .... .... .... .... ...1 .... .... = S_IRUSR: Set
            .... .... .... .... .... .... 1... .... = S_IWUSR: Set
            .... .... .... .... .... .... .0.. .... = S_IXUSR: Not set
            .... .... .... .... .... .... ..1. .... = S_IRGRP: Set
            .... .... .... .... .... .... ...0 .... = S_IWGRP: Not set
            .... .... .... .... .... .... .... 0... = S_IXGRP: Not set
            .... .... .... .... .... .... .... .0.. = S_IROTH: Not set
            .... .... .... .... .... .... .... ..0. = S_IWOTH: Not set
            .... .... .... .... .... .... .... ...0 = S_IXOTH: Not set
        Number of hard links: 1
        UID: 0
        GID: 0
        Root device: 0x0000000000000000
        Size: 2684354560
        Block size: 4096
        Blocks: 5242880
        Time of last access: Apr 28, 2016 17:58:23.604815960 Hora de verano romance
        Time of last modification: Apr 28, 2016 18:00:33.873544532 Hora de verano romance
        Time of last status change: Apr 28, 2016 18:00:33.874544506 Hora de verano romance
    Size: 32768
    Dict, contains 1 item
        [Size: 38 (40 bytes inc. RPC-roundup)]
        Items: 1
        glusterfs.inodelk-count: 1
        [RPC-roundup bytes: 00005c5c5c5c5c5c5c5c0000000000000000000000000000...]
    Data (32768 bytes)
        <raw data stripped>
        Data: 5c5c5c5c5c5c5c5c00000000000000000000000000000000...
        [Length: 32768]

We can see the answer has been reassembled from 2 TCP segments, the first one being of 116 bytes. Probably this is the cause of the problem.

Maybe the XDR decoding should make use of additional data in the second iovec entry if needed ?

Comment 1 Vijay Bellur 2016-05-11 11:15:59 UTC

REVIEW: http://review.gluster.org/14293 (socket: Fix incorrect handling of partial reads) posted (#1) for review on release-3.8 by Xavier Hernandez (xhernandez)

Comment 2 Vijay Bellur 2016-05-12 05:17:30 UTC

COMMIT: http://review.gluster.org/14293 committed in release-3.8 by Raghavendra G (rgowdapp) 
------
commit 11dca385555ae086f14bb26a33ec9713b5d09ee6
Author: Xavier Hernandez <xhernandez>
Date:   Mon May 9 12:50:39 2016 +0200

    socket: Fix incorrect handling of partial reads
    
    The usage of function local variables in the protocol state
    machine caused an incorrect behaviour when a partial read
    from the socket forced the function to return and restart
    later when more data was available. At this point the local
    variables contained incorrect data.
    
    > Change-Id: I4db1f4ef5c46a3d2d7f7c5328e906188c3af49e6
    > BUG: 1334285
    > Signed-off-by: Xavier Hernandez <xhernandez>
    > Reviewed-on: http://review.gluster.org/14270
    > Reviewed-by: Raghavendra G <rgowdapp>
    > Smoke: Gluster Build System <jenkins.com>
    > NetBSD-regression: NetBSD Build System <jenkins.org>
    > Tested-by: Raghavendra G <rgowdapp>
    > CentOS-regression: Gluster Build System <jenkins.com>
    
    Change-Id: I92b7c91b4c0dfc15224aea39308c93b27028dd4f
    BUG: 1334287
    Signed-off-by: Xavier Hernandez <xhernandez>
    Reviewed-on: http://review.gluster.org/14293
    NetBSD-regression: NetBSD Build System <jenkins.org>
    Reviewed-by: Raghavendra G <rgowdapp>
    CentOS-regression: Gluster Build System <jenkins.com>
    Smoke: Gluster Build System <jenkins.com>

Comment 3 Niels de Vos 2016-06-16 14:05:31 UTC

This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.8.0, please open a new bug report.

glusterfs-3.8.0 has been announced on the Gluster mailinglists [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution.

[1] http://blog.gluster.org/2016/06/glusterfs-3-8-released/
[2] http://thread.gmane.org/gmane.comp.file-systems.gluster.user