Until recently, when I pressed Option–Cmd–Eject, my MacBook Pro used to go to sleep almost immediately (e.g. within three or four seconds). It's a MBP 8,2 with an SSD and 8GB RAM.
Since a week or so, it takes much longer; around fifteen seconds or even more. Here's the console output from pressing the keyboard shortcut until the kernel's sleep
message:
7/31/11 11:50:59 PM mDNSResponder[17] Client application bug: DNSServiceResolve(charon._smb._tcp.local.) active for over two minutes. This places considerable burden on the network.
7/31/11 11:50:59 PM mDNSResponder[17] Client application bug: DNSServiceResolve(Meredith(SSH)._ssh._tcp.local.) active for over two minutes. This places considerable burden on the network.
7/31/11 11:51:03 PM kernel PM notification timeout (pid 3715, Google Chrome He)
7/31/11 11:51:03 PM kernel hibernate image path: /var/vm/sleepimage
7/31/11 11:51:03 PM kernel sizeof(IOHibernateImageHeader) == 512
7/31/11 11:51:03 PM kernel Opened file /var/vm/sleepimage, size 8589934592, partition base 0xc805000, maxio 400000 ssd 1
7/31/11 11:51:03 PM kernel hibernate image major 14, minor 2, blocksize 512, pollers 5
7/31/11 11:51:03 PM kernel hibernate_alloc_pages flags 00000000, gobbling 0 pages
7/31/11 11:51:03 PM kernel hibernate_setup(0) took 0 ms
7/31/11 11:51:06 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
7/31/11 11:51:06 PM kernel System SafeSleep
7/31/11 11:51:15 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0x00000000
7/31/11 11:51:15 PM kernel hibernate_page_list_setall start 0xffffff809225b000, 0xffffff809229b000
7/31/11 11:51:15 PM kernel hibernate_page_list_setall time: 447 ms
7/31/11 11:51:15 PM kernel pages 1564968, wire 250981, act 375476, inact 548, spec 267, zf 6140, throt 0, could discard act 115301 inact 538951 purgeable 39710 spec 237594
7/31/11 11:51:15 PM kernel hibernate_page_list_setall found pageCount 633412
7/31/11 11:51:15 PM kernel IOHibernatePollerOpen, ml_get_interrupts_enabled 0
7/31/11 11:51:15 PM kernel IOHibernatePollerOpen(0)
7/31/11 11:51:15 PM kernel writing 631584 pages
7/31/11 11:51:15 PM kernel encryptStart 69aa50
7/31/11 11:51:15 PM kernel encryptEnd 89ade70
7/31/11 11:51:15 PM kernel image1Size 310265344, encryptStart1 69aa50, End1 89ade70
7/31/11 11:51:15 PM kernel encryptStart 127e4600
7/31/11 11:51:15 PM kernel encryptEnd 40c9f7c0
7/31/11 11:51:15 PM kernel PMStats: Hibernate write took 8066 ms
7/31/11 11:51:15 PM kernel all time: 8066 ms, comp time: 2645 ms, deco time: 0 ms,
7/31/11 11:51:15 PM kernel image 1086978048, uncompressed 2587336704 (631674), compressed 1074427632 (41%), sum1 4849c5df, sum2 d1708974
7/31/11 11:51:15 PM kernel wired_pages_encrypted 146094, wired_pages_clear 103149, dirty_pages_encrypted 382431
7/31/11 11:51:15 PM kernel hibernate_write_image done(0)
7/31/11 11:51:15 PM kernel sleep
Note that I am aware of this question, but I haven't changed anything in the system and it used to go to sleep faster than now. Considering that PMStats
reports a hibernation write time of over eight seconds, this leads me into thinking it is the RAM content.
The only system changes I remember are installing the two updates for 10.6 that came out just before Lion's final release. Did that change anything?
How would I go about troubleshooting this?
Even after
sudo pmset -a hibernatemode 0
sudo pmset -a hibernatemode 1
it takes roughly the same time to sleep.
FYI, here's the information from a "clean" (non-safe) sleep:
8/1/11 2:26:49 PM kernel AirPort: Link Down on en1. Reason 8 (Disassociated because station leaving).
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0x00000000
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:50 PM configd[14] network configuration changed.
8/1/11 2:26:50 PM kernel System Sleep
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0x00000000
8/1/11 2:26:50 PM configd[14] network configuration changed.
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:50 PM configd[14] setting hostname to "charon.local"
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:50 PM kernel IOThunderboltSwitch::i2cWriteDWord - status = 0xe00002ed
8/1/11 2:26:58 PM kernel Wake reason = EHC2
8/1/11 2:26:58 PM kernel System Wake
Best Answer
Exactly. Your RAM contents are written to the disk, in the default sleep mode, called "Safe sleep", which takes longer the more RAM you have. You can modify this with the PMSet command in terminal.
and then reboot.
If you disable this option, you can also remove the sleepimage file from
/var/vm
, and reclaim 8GB of hard disk space. Do note that you will lose data then if you sleep for too long and the battery dies.