kernel-uek-4.1.2.124.16.2.el7uek halts on boot
Clash Royale CLAN TAG#URR8PPP
up vote
0
down vote
favorite
I had a bunch of Oracle Linux 7.5 VMs running on VMware ESXi patch and reboot overnight. They all halted right after displaying their disk lists on the console, for however many disks they have:
[ s.mmmuuu] sd 1:0:0:0: [sda]: Assuming drive cache: write through
About half of the hosts had an additional line after the disk list:
[ s.mmmuuu] dracut-pre-trigger[278]: modprobe: ERROR: could not insert 'camellia_aesni_av2': No such device
Similar messages appear in other console text streams without halting the kernel, so I don't think the lack of the camellia ciphers is the specific problem.
Unpatched OL7 hosts running kernel-uek-4.1.12.124.14.3.el7uek.x86_64 do not halt during boot. The current workaround is to restart the hosts with kernel-3.10.0-862.3.2.el7.x86_64. I don't see anything in Oracle Linux documentation or on Google about this, yet. I'm also not convinced this is a kernel problem more than a boot loader problem. More to come...
UPDATE 1, 2018-06-18
Booting without rhgb quiet shows the hang following a different crypto algorithm check.
[ 1.545...] alg: self-tests for ansi_cprng (stdrng) passed
[ 1.545...] alg: No test for fips(ansi_cprng) (fips_ansi_cprng)
[ 1.563...] AVX2 or AES-NI instructions are not detected.
[ 1.464...] dracut-pre-trigger[266]: modprobe: ERROR: could not insert 'camellia_aesni_avx2': No such device
[ 1.684...] alg: self-tests for deflate-generic (deflate) passed
[ 1.687...] alg: self-tests for des3_ede-generic (des3_ede) passed
[ 1.611...] alg: self-tests for des3_ede-asm (des3_ede) passed
[ 1.612...] alg: self-tests for ecb-des3_ede-asm (ecb(des3_ede)) passed
And there it hung....
Trying again with rhgb quiet removed and fips=0 instead of fips=1 results in a good boot. Because, of course hanging at crypto algorithm test makes this a FIPS thing...
Taking another pass through "7.2 Enabling FIPS Mode on Oracle Linux" revealed the option for a package called dracut-fips-aesni, since our VMware hypervisors support AES. Installing that package and re-running the instructions in Chapter 7.2 just caused the hang to occur a couple lines earlier, at des3_ede-generic.
I went looking through /var/log/messages. The boot attempts that hang don't get far enough to record anything. I'll get the crypto messages from the previous UEKr4 kernel if I boot to that. I don't get anything from the current UEKr4 kernel + fips=1. I don't get anything from fips=0, and I don't get anything from bootking the UEKr3 kernel.
boot linux-kernel oracle-linux
edited Jul 4 at 8:52
muru
33.2k576140
asked Jun 14 at 15:46
dafydd
6302820
add a comment |Â
up vote
0
down vote
favorite
I had a bunch of Oracle Linux 7.5 VMs running on VMware ESXi patch and reboot overnight. They all halted right after displaying their disk lists on the console, for however many disks they have:
[ s.mmmuuu] sd 1:0:0:0: [sda]: Assuming drive cache: write through
About half of the hosts had an additional line after the disk list:
[ s.mmmuuu] dracut-pre-trigger[278]: modprobe: ERROR: could not insert 'camellia_aesni_av2': No such device
Similar messages appear in other console text streams without halting the kernel, so I don't think the lack of the camellia ciphers is the specific problem.
Unpatched OL7 hosts running kernel-uek-4.1.12.124.14.3.el7uek.x86_64 do not halt during boot. The current workaround is to restart the hosts with kernel-3.10.0-862.3.2.el7.x86_64. I don't see anything in Oracle Linux documentation or on Google about this, yet. I'm also not convinced this is a kernel problem more than a boot loader problem. More to come...
UPDATE 1, 2018-06-18
Booting without rhgb quiet shows the hang following a different crypto algorithm check.
[ 1.545...] alg: self-tests for ansi_cprng (stdrng) passed
[ 1.545...] alg: No test for fips(ansi_cprng) (fips_ansi_cprng)
[ 1.563...] AVX2 or AES-NI instructions are not detected.
[ 1.464...] dracut-pre-trigger[266]: modprobe: ERROR: could not insert 'camellia_aesni_avx2': No such device
[ 1.684...] alg: self-tests for deflate-generic (deflate) passed
[ 1.687...] alg: self-tests for des3_ede-generic (des3_ede) passed
[ 1.611...] alg: self-tests for des3_ede-asm (des3_ede) passed
[ 1.612...] alg: self-tests for ecb-des3_ede-asm (ecb(des3_ede)) passed
And there it hung....
Trying again with rhgb quiet removed and fips=0 instead of fips=1 results in a good boot. Because, of course hanging at crypto algorithm test makes this a FIPS thing...
Taking another pass through "7.2 Enabling FIPS Mode on Oracle Linux" revealed the option for a package called dracut-fips-aesni, since our VMware hypervisors support AES. Installing that package and re-running the instructions in Chapter 7.2 just caused the hang to occur a couple lines earlier, at des3_ede-generic.
I went looking through /var/log/messages. The boot attempts that hang don't get far enough to record anything. I'll get the crypto messages from the previous UEKr4 kernel if I boot to that. I don't get anything from the current UEKr4 kernel + fips=1. I don't get anything from fips=0, and I don't get anything from bootking the UEKr3 kernel.
boot linux-kernel oracle-linux
edited Jul 4 at 8:52
muru
33.2k576140
asked Jun 14 at 15:46
dafydd
6302820
Try removing therhgb quietlines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.
– wurtel
Jun 15 at 13:48
add a comment |Â
up vote
0
down vote
favorite
up vote
0
down vote
favorite
I had a bunch of Oracle Linux 7.5 VMs running on VMware ESXi patch and reboot overnight. They all halted right after displaying their disk lists on the console, for however many disks they have:
[ s.mmmuuu] sd 1:0:0:0: [sda]: Assuming drive cache: write through
About half of the hosts had an additional line after the disk list:
[ s.mmmuuu] dracut-pre-trigger[278]: modprobe: ERROR: could not insert 'camellia_aesni_av2': No such device
Similar messages appear in other console text streams without halting the kernel, so I don't think the lack of the camellia ciphers is the specific problem.
Unpatched OL7 hosts running kernel-uek-4.1.12.124.14.3.el7uek.x86_64 do not halt during boot. The current workaround is to restart the hosts with kernel-3.10.0-862.3.2.el7.x86_64. I don't see anything in Oracle Linux documentation or on Google about this, yet. I'm also not convinced this is a kernel problem more than a boot loader problem. More to come...
UPDATE 1, 2018-06-18
Booting without rhgb quiet shows the hang following a different crypto algorithm check.
[ 1.545...] alg: self-tests for ansi_cprng (stdrng) passed
[ 1.545...] alg: No test for fips(ansi_cprng) (fips_ansi_cprng)
[ 1.563...] AVX2 or AES-NI instructions are not detected.
[ 1.464...] dracut-pre-trigger[266]: modprobe: ERROR: could not insert 'camellia_aesni_avx2': No such device
[ 1.684...] alg: self-tests for deflate-generic (deflate) passed
[ 1.687...] alg: self-tests for des3_ede-generic (des3_ede) passed
[ 1.611...] alg: self-tests for des3_ede-asm (des3_ede) passed
[ 1.612...] alg: self-tests for ecb-des3_ede-asm (ecb(des3_ede)) passed
And there it hung....
Trying again with rhgb quiet removed and fips=0 instead of fips=1 results in a good boot. Because, of course hanging at crypto algorithm test makes this a FIPS thing...
Taking another pass through "7.2 Enabling FIPS Mode on Oracle Linux" revealed the option for a package called dracut-fips-aesni, since our VMware hypervisors support AES. Installing that package and re-running the instructions in Chapter 7.2 just caused the hang to occur a couple lines earlier, at des3_ede-generic.
I went looking through /var/log/messages. The boot attempts that hang don't get far enough to record anything. I'll get the crypto messages from the previous UEKr4 kernel if I boot to that. I don't get anything from the current UEKr4 kernel + fips=1. I don't get anything from fips=0, and I don't get anything from bootking the UEKr3 kernel.
boot linux-kernel oracle-linux
edited Jul 4 at 8:52
muru
33.2k576140
asked Jun 14 at 15:46
dafydd
6302820
I had a bunch of Oracle Linux 7.5 VMs running on VMware ESXi patch and reboot overnight. They all halted right after displaying their disk lists on the console, for however many disks they have:
[ s.mmmuuu] sd 1:0:0:0: [sda]: Assuming drive cache: write through
About half of the hosts had an additional line after the disk list:
[ s.mmmuuu] dracut-pre-trigger[278]: modprobe: ERROR: could not insert 'camellia_aesni_av2': No such device
Similar messages appear in other console text streams without halting the kernel, so I don't think the lack of the camellia ciphers is the specific problem.
Unpatched OL7 hosts running kernel-uek-4.1.12.124.14.3.el7uek.x86_64 do not halt during boot. The current workaround is to restart the hosts with kernel-3.10.0-862.3.2.el7.x86_64. I don't see anything in Oracle Linux documentation or on Google about this, yet. I'm also not convinced this is a kernel problem more than a boot loader problem. More to come...
UPDATE 1, 2018-06-18
Booting without rhgb quiet shows the hang following a different crypto algorithm check.
[ 1.545...] alg: self-tests for ansi_cprng (stdrng) passed
[ 1.545...] alg: No test for fips(ansi_cprng) (fips_ansi_cprng)
[ 1.563...] AVX2 or AES-NI instructions are not detected.
[ 1.464...] dracut-pre-trigger[266]: modprobe: ERROR: could not insert 'camellia_aesni_avx2': No such device
[ 1.684...] alg: self-tests for deflate-generic (deflate) passed
[ 1.687...] alg: self-tests for des3_ede-generic (des3_ede) passed
[ 1.611...] alg: self-tests for des3_ede-asm (des3_ede) passed
[ 1.612...] alg: self-tests for ecb-des3_ede-asm (ecb(des3_ede)) passed
And there it hung....
Trying again with rhgb quiet removed and fips=0 instead of fips=1 results in a good boot. Because, of course hanging at crypto algorithm test makes this a FIPS thing...
Taking another pass through "7.2 Enabling FIPS Mode on Oracle Linux" revealed the option for a package called dracut-fips-aesni, since our VMware hypervisors support AES. Installing that package and re-running the instructions in Chapter 7.2 just caused the hang to occur a couple lines earlier, at des3_ede-generic.
I went looking through /var/log/messages. The boot attempts that hang don't get far enough to record anything. I'll get the crypto messages from the previous UEKr4 kernel if I boot to that. I don't get anything from the current UEKr4 kernel + fips=1. I don't get anything from fips=0, and I don't get anything from bootking the UEKr3 kernel.
boot linux-kernel oracle-linux
edited Jul 4 at 8:52
muru
33.2k576140
asked Jun 14 at 15:46
dafydd
6302820
edited Jul 4 at 8:52
muru
33.2k576140
edited Jul 4 at 8:52
muru
33.2k576140
edited Jul 4 at 8:52
muru
33.2k576140
33.2k576140
asked Jun 14 at 15:46
dafydd
6302820
asked Jun 14 at 15:46
dafydd
6302820
asked Jun 14 at 15:46
dafydd
6302820
6302820
Try removing therhgb quietlines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.
– wurtel
Jun 15 at 13:48
add a comment |Â
Try removing therhgb quietlines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.
– wurtel
Jun 15 at 13:48
Try removing the
rhgb quiet lines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.– wurtel
Jun 15 at 13:48
Try removing the
rhgb quiet lines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.– wurtel
Jun 15 at 13:48
add a comment |Â
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2funix.stackexchange.com%2fquestions%2f449844%2fkernel-uek-4-1-2-124-16-2-el7uek-halts-on-boot%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Try removing the
rhgb quietlines from the kernel boot line to see much more detail about the boot process. I've never understood why one would prefer a "pretty" boot screen instead of an informative one.– wurtel
Jun 15 at 13:48