To cut a long story short, I've got two Fedora boxes both running Core 5. One is a Poweredge 1750 with 4GB RAM (MySQL master) and the other is an SC1425 with 1GB RAM and 512MB of swap space (MySQL slave).
The slave was running happily for days, but last night, the box went down. Now, the box will boot fine, but if I start the slave, it eats all the memory and crashes.
Slave's my.cnf memory settings
Code:
# The MySQL server
[mysqld]
port = 3311
socket = /tmp/mysql_slave1.sock
skip-locking
key_buffer = 256M
max_allowed_packet = 4M
table_cache = 256
sort_buffer_size = 1M
read_buffer_size = 1M
read_rnd_buffer_size = 4M
myisam_sort_buffer_size = 64M
thread_cache_size = 8
query_cache_size= 16M
# Try number of CPU's*2 for thread_concurrency
thread_concurrency = 8
Nothing in the mysql error log when it crashes. From /var/log/messages though:
Code:
Oct 2 11:09:57 h100465 kernel: Call Trace: <ffffffff802b5ce3>{out_of_memory+51} <ffffffff802677b6>{thread_return+0}
Oct 2 11:09:57 h100465 kernel: <ffffffff8020f13e>{__alloc_pages+594} <ffffffff8021242f>{__do_page_cache_readahead+169}
Oct 2 11:09:57 h100465 kernel: <ffffffff802299e5>{sync_page+0} <ffffffff80267b43>{io_schedule+50}
Oct 2 11:09:57 h100465 kernel: <ffffffff802687db>{__wait_on_bit_lock+91} <ffffffff880b5c2e>{:dm_mod:dm_any_congested+59}
Oct 2 11:09:57 h100465 kernel: <ffffffff80213e49>{filemap_nopage+326} <ffffffff80208825>{__handle_mm_fault+894}
Oct 2 11:09:57 h100465 kernel: <ffffffff80269d8e>{_spin_lock_irqsave+9} <ffffffff8026bc01>{do_page_fault+1120}
Oct 2 11:09:57 h100465 kernel: <ffffffff8022c286>{sys_newfstat+32} <ffffffff802638a5>{error_exit+0}
Oct 2 11:09:57 h100465 kernel: Mem-info:
Oct 2 11:09:57 h100465 kernel: Node 0 DMA per-cpu:
Oct 2 11:09:57 h100465 kernel: cpu 0 hot: high 0, batch 1 used:0
Oct 2 11:09:57 h100465 kernel: cpu 0 cold: high 0, batch 1 used:0
Oct 2 11:10:04 h100465 kernel: cpu 1 hot: high 0, batch 1 used:0
Oct 2 11:10:19 h100465 kernel: cpu 1 cold: high 0, batch 1 used:0
Oct 2 11:10:23 h100465 kernel: Node 0 DMA32 per-cpu:
Oct 2 11:10:23 h100465 kernel: cpu 0 hot: high 186, batch 31 used:42
Oct 2 11:10:24 h100465 kernel: cpu 0 cold: high 62, batch 15 used:59
Oct 2 11:10:24 h100465 kernel: cpu 1 hot: high 186, batch 31 used:113
Oct 2 11:10:25 h100465 kernel: cpu 1 cold: high 62, batch 15 used:21
Oct 2 11:10:28 h100465 kernel: Node 0 Normal per-cpu: empty
Oct 2 11:10:30 h100465 kernel: Node 0 HighMem per-cpu: empty
Oct 2 11:10:30 h100465 kernel: Free pages: 8072kB (0kB HighMem)
Oct 2 11:10:31 h100465 kernel: Active:122168 inactive:120990 dirty:0 writeback:0 unstable:0 free:2018 slab:4009 mapped:243215 pagetables:1946
Oct 2 11:10:32 h100465 kernel: Node 0 DMA free:4004kB min:36kB low:44kB high:52kB active:3252kB inactive:2612kB present:9544kB pages_scanned:5997 all_unrecla
imable? yes
Oct 2 11:10:33 h100465 kernel: lowmem_reserve[]: 0 992 992 992
Oct 2 11:10:34 h100465 kernel: Node 0 DMA32 free:4068kB min:4008kB low:5008kB high:6012kB active:485420kB inactive:481348kB present:1015812kB pages_scanned:2
0768 all_unreclaimable? no
Oct 2 11:10:35 h100465 kernel: lowmem_reserve[]: 0 0 0 0
Oct 2 11:10:35 h100465 kernel: Node 0 Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
Oct 2 11:10:35 h100465 kernel: lowmem_reserve[]: 0 0 0 0
Oct 2 11:10:36 h100465 kernel: Node 0 HighMem free:0kB min:128kB low:128kB high:128kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable?
no
Oct 2 11:10:36 h100465 kernel: lowmem_reserve[]: 0 0 0 0
Oct 2 11:10:37 h100465 kernel: Node 0 DMA: 1*4kB 2*8kB 1*16kB 0*32kB 0*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 4004kB
Oct 2 11:10:37 h100465 kernel: Node 0 DMA32: 15*4kB 9*8kB 6*16kB 0*32kB 0*64kB 0*128kB 1*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 4068kB
Oct 2 11:10:38 h100465 kernel: Node 0 Normal: empty
Oct 2 11:10:39 h100465 kernel: Node 0 HighMem: empty
Oct 2 11:10:39 h100465 kernel: Swap cache: add 134000, delete 133992, find 393/709, race 0+0
Oct 2 11:10:40 h100465 kernel: Free swap = 0kB
Oct 2 11:10:40 h100465 kernel: Total swap = 524280kB
Oct 2 11:10:40 h100465 kernel: Free swap: 0kB
Oct 2 11:10:41 h100465 kernel: 262080 pages of RAM
Oct 2 11:10:41 h100465 kernel: 6152 reserved pages
Oct 2 11:10:42 h100465 kernel: 4459 pages shared
Oct 2 11:10:42 h100465 kernel: 8 pages swap cached
Oct 2 11:10:42 h100465 kernel: Out of Memory: Kill process 3403 (mysqld) score 404207 and children.
Oct 2 11:10:42 h100465 kernel: Out of memory: Killed process 3403 (mysqld).
So - is this a fault with MySQL or Fedora? Any suggestions as to how to proceed from here?!
Thanks in advance...
Neil.
Linear Mode
