OS Event log - messages

Oct 14 07:09:07 Hostname kernel: ACPI Error: SMBus/IPMI/GenericSerialBus write requires Buffer of length 66, found length 32 (20130517/exfield-389)

Oct 14 07:09:07 Hostname kernel: ACPI Error: Method parse/execution failed [\_SB_.PMI0._PMM] (Node ffff88016d9e8ed8), AE_AML_BUFFER_LIMIT (20130517/psparse-536)

Oct 14 07:09:07 Hostname kernel: ACPI Exception: AE_AML_BUFFER_LIMIT, Evaluating _PMM (20130517/power_meter-339)


or


May 10 10:14:56 Hostname kernel: ACPI Error: SMBus or IPMI write requires Buffer of length 66, found length 32 (20090903/exfield-286)

May 10 10:14:56 Hostname kernel: ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PMI0._GHL] (Node ffff881074e41d08), AE_AML_BUFFER_LIMIT

May 10 10:14:56 Hostname kernel: ACPI Exception: AE_AML_BUFFER_LIMIT, Evaluating _GHL (20090903/power_meter-204)



Why do I see "ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PMI0._GHL] (Node ffff8101023768f0), AE_SUPPORT" when my RHEL 5 system boots?
https://access.redhat.com/site/solutions/35475

Resolution
No action is required.  These messages cannot be removed at the current time.


ACPI Error(ACPI Error: SMBus or IPMI write requires Buffer of length X, found length Y)
https://access.redhat.com/site/solutions/48109

Resolution
This message can be safely ignored.


Advisory: ProLiant G6 and G7 Servers - Advanced Configuration and Power Interface (ACPI) Errors May Be Reported by Red Hat Enterprise Linux 6 and SUSE Linux Enterprise Server 11 SP1 During Startup

https://support.hpe.com/hpesc/public/docDisplay?docId=c02642731&docLocale=en_US

Because the System ROM ACPI interface causing these error messages is not used in the Linux operating system environment, these messages are benign and can be safely ignored.




Posted by 스쳐가는인연

댓글을 달아 주세요

HPE NIC driver update through source rpm file from Ubuntu OS.

 

Test System
DL380 Gen10+562SFP+(X710 chipset)

 

1. Install SPP2020.03
2. Set WP to Virtualization - Max Performance.
3. Install Ubuntu 18.04.5 LTS

 

Check inbox driver (dmesg | grep i40e)
[ 3.733322] i40e: Intel(R) Ethernet Connection XL710 Network Driver - version 2.1.14-k

 

a. enable root account

# sudo passwd root

 

b. enable root login via ssh

# vim /etc/ssh/sshd_config

PermitRootLogin yes

 

# service ssh restart

 

4.1 Install rpm for use rpm package on Ubuntu

# apt install rpm

# rpm -ivh ./hp-i40e-2.10.19.30-2.all.src.rpm

rpm: RPM should not be used directly install RPM packages, use Alien instead!
rpm: However assuming you know what you are doing...
Updating / installing...
1:hp-i40e-2.10.19.30-2.all ################################# [100%]

 

rpm -ql ./hp-i40e-2.10.19.30-2.all.src.rpm

hp-i40e-2.10.19.30.tar.gz
hp-i40e.files
hp-i40e.spec

 

>> package installed at

/rpmbuild/SOURCE

 

4.2. Install alien for use rpm package on Ubuntu

# apt install alien

# alien hp-i40e-2.10.19.30-2.all.src.rpm

hp-i40e_2.10.19.30-3_amd64.deb generated

 

# dpkg -i hp-i40e_2.10.19.30-3_amd64.deb

 

# dpkg -L hp-i40e

/.
/hp-i40e-2.10.19.30.tar.gz
/hp-i40e.files
/hp-i40e.spec
/usr
/usr/share
/usr/share/doc
/usr/share/doc/hp-i40e
/usr/share/doc/hp-i40e/changelog.Debian.gz
/usr/share/doc/hp-i40e/copyright

 

>> package installed at

/

 

5. Move source package to temporary space

# mkdir /tmp/i40e

# mv hp-i40e* /tmp/i40e

 

6. Install new driver.

# tar zxf hp-i40e-2.10.19.30.tar.gz

# cd /tmp/i40e/hp-i40e-2.10.19.30/src

# make install

 

7. remove old driver and then load new driver

# rmmod i40e; modprobe i40e

 

8. update. seems automatically done during install but. confirm

# update-initramfs -u

 

Reboot and check new driver loaded.

[ 4.028479] i40e: Intel(R) 40-10 Gigabit Ethernet Connection Network Driver - version 2.10.19.30

Posted by 스쳐가는인연

댓글을 달아 주세요

Directory copy

# cp -r source_directory target_directory

 

Multiple file overwrite without interactive

# \cp -rf source_directory target_directory
# /bin/cp -rf source_directory target_directory

 

출처:

cp 강제로 덮어쓰기
https://zetawiki.com/wiki/CentOS_cp_%EA%B0%95%EC%A0%9C%EB%A1%9C_%EB%8D%AE%EC%96%B4%EC%93%B0%EA%B8%B0

리눅스에서 폴더 통째로 복사하기
https://zetawiki.com/wiki/%EB%A6%AC%EB%88%85%EC%8A%A4%EC%97%90%EC%84%9C_%ED%8F%B4%EB%8D%94_%ED%86%B5%EC%A7%B8%EB%A1%9C_%EB%B3%B5%EC%82%AC%ED%95%98%EA%B8%B0

Posted by 스쳐가는인연

댓글을 달아 주세요

Summary

fdisk:

# fdisk /dev/sdb     #추가된 Disk
n                         #파티션 생성
w                        #변경사항 기록

e.g.) /dev/sdb의 partition 1 > /dev/sdb1

# pvcreate /dev/sdb1                              #물리볼륨 생성
# vgcreate vgdata1 /dev/sdb1                   #볼륨그룹 생성
# lvcreate -n lvdata1 -l 100%FREE vgdata1   #논리볼륨 생성

# mkfs.ext4 /dev/vgdata1/lvdata1                  #파일시스템 생성

# mkdir /Depot                                         #마운트포인트 생성
# mount -t ext4 /dev/vgdata1/lvdata1 /Depot  #마운트

# df -l                    #상태 확인


gdisk:
------------------------
# gdisk /dev/sdc
p       << print partition table
n
W
------------------------


parted:
# parted /dev/sdc
------------------------
# parted
# select /dev/sdc
------------------------

------------------------
# mktable gpt  << gpt
------------------------
# mktable msdos << mbr
------------------------

------------------------
# mkpart primary ext4 0% 400GB
# mkpart primary ext4 0% 100%
------------------------

# align-check optimal 1

# quit

# partprobe /dev/sdc

# mkfs.ext4 -L <%PartitionName-Label%> /dev/sdc1
# mkfs.ext4 -L 900GB /dev/sdc1
# e2label /dev/sda1 <%newlabel%>

# mkdir -p /900GB
# mount -o defaults /dev/sdc1 /900GB

#cat /etc/fstab
--------------------------
## select one method
/dev/sdc1 /900GB ext4 defaults 0 2
# UUID=4b313333-a7b5-48c1-a957-d77d637e4fda /mnt/data ext4 defaults 0 2
# LABEL=900GB /900GB ext4 defaults 0 2
--------------------------


참조:
https://www.certdepot.net/rhel7-create-delete-partitions/
https://www.digitalocean.com/community/tutorials/how-to-partition-and-format-storage-devices-in-linux

Posted by 스쳐가는인연

댓글을 달아 주세요

e.g.) IO workload

 

1. Create Job - 1G file create 100ea per every 20 mins / then remove it.

# crontab -u root crontab

# crontab -e
*/20 * * * * for i in {1..100}; do dd if=/dev/zero of=/tmp/dummy1G-$i bs=1 count=0 seek=1G; done
*/20 * * * * rm -f /tmp/dummy1G-*

 

2. List current cron job

# crontab -l

 

3. Remove cron job

# crontab -r

 

4. Repeat per second

* * * * * * <%command%> & sleep 30; <%command%>

 

5. restart cron daemon

service crond restart

 

cf. 

1 * * * *, not per every minute, it means every hours 1st minute. 1:01, 2:01, ...

* * * * *, execute per every minute.

* 1 * * *, not every hours, it means every days 1 o'clock.

01 * * * *, execute per every hour

 

 

참조.

https://greenhappy.tistory.com/entry/%EB%A6%AC%EB%88%85%EC%8A%A4-crontab-%EC%B4%88%EB%8B%A8%EC%9C%84-%EC%8B%A4%ED%96%89-%ED%95%98%EA%B8%B0

 

https://zetawiki.com/wiki/%ED%81%AC%EB%A1%A0%ED%83%AD_%EC%9E%91%EC%97%85_5%EB%B6%84%EB%A7%88%EB%8B%A4_%EC%88%98%ED%96%89

 

https://soooprmx.com/archives/6786

Posted by 스쳐가는인연

댓글을 달아 주세요