FAQ
Hello all,

Redis had slowed down quite a bit. I have given the output of INFO command
below. There are sets on this redis instance. RAM is 16GB. Is there
anything that seems suspicious in the following INFO output? What should be
the value of mem_fragmentation_ratio in order to suspect memory
fragmentation?

Also, the background save thread has been running since a minute. Size of
dump.rdb is 2GB and size of the temp.rdb is 675MB.

# Server
redis_version:2.9.5
redis_git_sha1:6aa2f989
redis_git_dirty:0
arch_bits:64
multiplexing_api:epoll
gcc_version:4.3.3
process_id:28606
run_id:0bcebead93bdb3ddebb3d606e68d240f7637edc5
tcp_port:6479
uptime_in_seconds:93728
uptime_in_days:1
lru_clock:146225

# Clients
connected_clients:11
client_longest_output_list:0
client_biggest_input_buf:1
blocked_clients:0

# Memory
used_memory:10888087120
used_memory_human:10.14G
used_memory_rss:5089050624
used_memory_peak:10957659624
used_memory_peak_human:10.21G
used_memory_lua:27648
mem_fragmentation_ratio:0.47
mem_allocator:jemalloc-2.2.5

# Persistence
loading:0
aof_enabled:0
changes_since_last_save:2292049
bgsave_in_progress:1
last_save_time:1364588622
last_bgsave_status:ok
bgrewriteaof_in_progress:0

# Stats
total_connections_received:21
total_commands_processed:775090319
instantaneous_ops_per_sec:0
rejected_connections:0
expired_keys:0
evicted_keys:0
keyspace_hits:7476565
keyspace_misses:9908481
pubsub_channels:0
pubsub_patterns:0
latest_fork_usec:494668

# Replication
role:master
connected_slaves:0

# CPU
used_cpu_sys:688.32
used_cpu_user:6836.11
used_cpu_sys_children:529.67
used_cpu_user_children:3776.61

# Cluster
cluster_enabled:0

# Keyspace
db0:keys=463,expires=0
db1:keys=10966165,expires=0
db2:keys=233,expires=0

Thanks in advance.

Regards,
Raghava.

--
You received this message because you are subscribed to the Google Groups "Redis DB" group.
To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+unsubscribe@googlegroups.com.
To post to this group, send email to redis-db@googlegroups.com.
Visit this group at http://groups.google.com/group/redis-db?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Search Discussions

  • Josiah Carlson at Mar 30, 2013 at 5:12 am
    Redis is using more memory than you have available in your system, so is
    swapped out to disk.

    Increase available memory or reduce Redis memory use, and Redis will be
    fast again.

    - Josiah

    On Fri, Mar 29, 2013 at 9:50 PM, Raghava Mutharaju wrote:

    Hello all,

    Redis had slowed down quite a bit. I have given the output of INFO command
    below. There are sets on this redis instance. RAM is 16GB. Is there
    anything that seems suspicious in the following INFO output? What should be
    the value of mem_fragmentation_ratio in order to suspect memory
    fragmentation?

    Also, the background save thread has been running since a minute. Size of
    dump.rdb is 2GB and size of the temp.rdb is 675MB.

    # Server
    redis_version:2.9.5
    redis_git_sha1:6aa2f989
    redis_git_dirty:0
    arch_bits:64
    multiplexing_api:epoll
    gcc_version:4.3.3
    process_id:28606
    run_id:0bcebead93bdb3ddebb3d606e68d240f7637edc5
    tcp_port:6479
    uptime_in_seconds:93728
    uptime_in_days:1
    lru_clock:146225

    # Clients
    connected_clients:11
    client_longest_output_list:0
    client_biggest_input_buf:1
    blocked_clients:0

    # Memory
    used_memory:10888087120
    used_memory_human:10.14G
    used_memory_rss:5089050624
    used_memory_peak:10957659624
    used_memory_peak_human:10.21G
    used_memory_lua:27648
    mem_fragmentation_ratio:0.47
    mem_allocator:jemalloc-2.2.5

    # Persistence
    loading:0
    aof_enabled:0
    changes_since_last_save:2292049
    bgsave_in_progress:1
    last_save_time:1364588622
    last_bgsave_status:ok
    bgrewriteaof_in_progress:0

    # Stats
    total_connections_received:21
    total_commands_processed:775090319
    instantaneous_ops_per_sec:0
    rejected_connections:0
    expired_keys:0
    evicted_keys:0
    keyspace_hits:7476565
    keyspace_misses:9908481
    pubsub_channels:0
    pubsub_patterns:0
    latest_fork_usec:494668

    # Replication
    role:master
    connected_slaves:0

    # CPU
    used_cpu_sys:688.32
    used_cpu_user:6836.11
    used_cpu_sys_children:529.67
    used_cpu_user_children:3776.61

    # Cluster
    cluster_enabled:0

    # Keyspace
    db0:keys=463,expires=0
    db1:keys=10966165,expires=0
    db2:keys=233,expires=0

    Thanks in advance.

    Regards,
    Raghava.

    --
    You received this message because you are subscribed to the Google Groups
    "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an
    email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.

    --
    You received this message because you are subscribed to the Google Groups "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.
  • Raghava Mutharaju at Mar 30, 2013 at 5:31 am
    Thank you for the reply. Based on which parameters did you arrive at that
    conclusion? Is it that for background save, at least 2X DB size (which is
    10GB in memory) is recommended?

    Regards,
    Raghava.

    On Sat, Mar 30, 2013 at 1:12 AM, Josiah Carlson wrote:

    Redis is using more memory than you have available in your system, so is
    swapped out to disk.

    Increase available memory or reduce Redis memory use, and Redis will be
    fast again.

    - Josiah


    On Fri, Mar 29, 2013 at 9:50 PM, Raghava Mutharaju <
    m.vijayaraghava@gmail.com> wrote:
    Hello all,

    Redis had slowed down quite a bit. I have given the output of INFO
    command below. There are sets on this redis instance. RAM is 16GB. Is there
    anything that seems suspicious in the following INFO output? What should be
    the value of mem_fragmentation_ratio in order to suspect memory
    fragmentation?

    Also, the background save thread has been running since a minute. Size of
    dump.rdb is 2GB and size of the temp.rdb is 675MB.

    # Server
    redis_version:2.9.5
    redis_git_sha1:6aa2f989
    redis_git_dirty:0
    arch_bits:64
    multiplexing_api:epoll
    gcc_version:4.3.3
    process_id:28606
    run_id:0bcebead93bdb3ddebb3d606e68d240f7637edc5
    tcp_port:6479
    uptime_in_seconds:93728
    uptime_in_days:1
    lru_clock:146225

    # Clients
    connected_clients:11
    client_longest_output_list:0
    client_biggest_input_buf:1
    blocked_clients:0

    # Memory
    used_memory:10888087120
    used_memory_human:10.14G
    used_memory_rss:5089050624
    used_memory_peak:10957659624
    used_memory_peak_human:10.21G
    used_memory_lua:27648
    mem_fragmentation_ratio:0.47
    mem_allocator:jemalloc-2.2.5

    # Persistence
    loading:0
    aof_enabled:0
    changes_since_last_save:2292049
    bgsave_in_progress:1
    last_save_time:1364588622
    last_bgsave_status:ok
    bgrewriteaof_in_progress:0

    # Stats
    total_connections_received:21
    total_commands_processed:775090319
    instantaneous_ops_per_sec:0
    rejected_connections:0
    expired_keys:0
    evicted_keys:0
    keyspace_hits:7476565
    keyspace_misses:9908481
    pubsub_channels:0
    pubsub_patterns:0
    latest_fork_usec:494668

    # Replication
    role:master
    connected_slaves:0

    # CPU
    used_cpu_sys:688.32
    used_cpu_user:6836.11
    used_cpu_sys_children:529.67
    used_cpu_user_children:3776.61

    # Cluster
    cluster_enabled:0

    # Keyspace
    db0:keys=463,expires=0
    db1:keys=10966165,expires=0
    db2:keys=233,expires=0

    Thanks in advance.

    Regards,
    Raghava.

    --
    You received this message because you are subscribed to the Google Groups
    "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an
    email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.

    --
    You received this message because you are subscribed to the Google Groups
    "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an
    email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.

    --
    You received this message because you are subscribed to the Google Groups "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.
  • Didier Spezia at Mar 30, 2013 at 10:07 am
    I guess this conclusion comes from the fact the used_memory_rss is half
    the used_memory,
    which concretely means half of Redis memory has been swapped out.

    This can be due to Redis itself at bgsave time (if your instance perform a
    lot of write operations),
    or any other process which takes memory on the same machine (or perform
    I/Os on large files).

    Regards,
    Didier.

    --
    You received this message because you are subscribed to the Google Groups "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.
  • Raghava Mutharaju at Mar 30, 2013 at 3:48 pm
    Hello Didier,

    I think its the bgsave which is taking the memory and that it turn leads to
    Redis using only part of the memory.

    Thank you for the replies.

    Regards,
    Raghava.


    On Sat, Mar 30, 2013 at 6:07 AM, Didier Spezia wrote:


    I guess this conclusion comes from the fact the used_memory_rss is half
    the used_memory,
    which concretely means half of Redis memory has been swapped out.

    This can be due to Redis itself at bgsave time (if your instance perform a
    lot of write operations),
    or any other process which takes memory on the same machine (or perform
    I/Os on large files).

    Regards,
    Didier.

    --
    You received this message because you are subscribed to the Google Groups
    "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an
    email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.

    --
    You received this message because you are subscribed to the Google Groups "Redis DB" group.
    To unsubscribe from this group and stop receiving emails from it, send an email to redis-db+unsubscribe@googlegroups.com.
    To post to this group, send email to redis-db@googlegroups.com.
    Visit this group at http://groups.google.com/group/redis-db?hl=en.
    For more options, visit https://groups.google.com/groups/opt_out.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupredis-db @
categoriesredis
postedMar 30, '13 at 4:51a
activeMar 30, '13 at 3:48p
posts5
users3
websiteredis.io
irc#redis

People

Translate

site design / logo © 2022 Grokbase