Thanks Mark,
In the meantime the DBA restarted the dataserver so I'll check those commands if this situation is raised again.
Then in the meantime I checked the dataserver log file and I noticed ugly message:
00:0007:00000:00008:2015/10/01 11:15:28.91 server An attempt was made to fetch logical page '61005824' for database 'DATABASE A' (5), object '' (99), index '<Unknown>' (0), partition '<Unknown>' (99) from cache 'default data cache'. Wrong logical page '60817408' was found in cache.
00:0006:00000:00172:2015/10/01 11:15:54.93 server Starting diagnostics for read failure:
00:0006:00000:00172:2015/10/01 11:15:54.93 server Device Information:
Device number = 14
Phyname = /data1/sybase/DATASERVER A/databases/devfs6
00:0006:00000:00172:2015/10/01 11:15:54.93 server Buffer Information:
Buf addr = 0x0x8fd709c28, Mass addr = 0x0x8fd709c28
Buf pageno = 61005824, Mass pageno = 61005824, dbid = 5
00:0006:00000:00172:2015/10/01 11:15:54.93 server
Buf virtpg = 61005824, Mass virtpg = 14114816
Buf stat = 0x1, Mass stat = 0x4001008
Mass keep = 1, Mass awaited = 0
I believe this error leads to an incorrect disconnection of SPID 172