PH18920: EDGE LOAD BALANCER V9.50.1 CRASH IN LBLL_SELECT1_NL WITH HIGH AVAILABILITY REPLICATION AND AFFINITY BASED FORWARDING

8.5.5.17: WebSphere Application Server V8.5.5 Fix Pack 17
9.0.5.3: WebSphere Application Server traditional Version 9.0.5 Fix Pack 3

APAR status

• Closed as program error.

Error description

• When using Edge Server Load balancer in high availability mode,
  the standby load balancer crashed and generated a core dump.
  High availability replication must be configured to replicate
  affinity records to experience this crash.
  
  Crash can occur on all supported operating systems.
  

Local fix

• N/A
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:  IBM WebSphere Application Server Load       *
  *                  Balancer for IPv4 and IPv6 users            *
  ****************************************************************
  * PROBLEM DESCRIPTION: Crash occurs on standby load balancer   *
  *                      when running with high availability     *
  *                      and affinity (or conn+affin)            *
  *                      replication.                            *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  The affinity table get corrupted and reference to a bad
  pointer occurs.
  Indicators can be seen prior to crash in the
  /var/logs/messages file(Linux) or ibmlb2.log (Windows) or
  ulb_logX* (AIX using lb_kext). Similar messagse will be
  observed:
  lb_ha.c:414 HA:add AT record failed
  

Problem conclusion

• Crash stack will look similar to:
  #0  lbll_select1_nl (fromlist=0x7f71639eb9c8,
  key=0x7f717300b7b0) at lb_linkedlist.c:496
  #1  lbll_select1 (fromlist=0x7f71639eb9c8, key=0x7f717300b7b0)
  at lb_linkedlist.c:475
  #2  lbht_select1 (fromtable=0x7f71638dd010,
  key=0x7f717300b7b0) at lb_hashtable.c:109
  #3  _lbha_handlereprecords (pkt=0x7f715c00093f) at lb_ha.c:369
  #4  _lbha_event (event=2, pkt=0x7f715c00093f) at lb_ha.c:674
  #5  lbhahb_pbuf_entry (pkt=0x7f712f74cfa0) at lb_ha.c:960
  #6  lbip_pbuf_entry (pkt=0x7f712f74cfa0) at lb_ip.c:188
  #7  lbipv4_entry (ethhdr=0x7f715c000915, iphdr=0x7f715c000923,
  bufflen=836, osopaque=0x7f715c0008c0) at lb_ipv4.c:138
  #8  lx_recv (psfd=4) at lx_entry.c:374
  #9  lx_recv_pthread (psfd_parm=0x55751c <psfd>) at
  lx_entry.c:245
  #10 start_thread () from /lib64/libpthread.so.0
  #11 clone () from /lib64/libc.so.6
  
  Corrected code to guard against corruption and bad pointer
  access under corrupted situations.
  Fix levels:
  8.5.5.17
  9.0.5.3
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  WS EDGE LB IPV4

• Fixed component ID

  5724H8812

Applicable component levels

• R850 PSY

     UP

• R900 PSY

     UP